From nih-image-d-request@io.ece.drexel.edu Tue Sep 28 06:18 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA22114; Tue, 28 Sep 1999 06:18:23 -0400 (EDT) Date: Tue, 28 Sep 1999 06:18:23 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199909281018.GAA22114@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #221 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/221 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 8084 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 221 Today's Topics: Focus macro (suite) [ Bertrand Menard ] Re: What means tokenized? [ Norbert Vischer To: nih-image@io.ece.drexel.edu Subject: Focus macro (suite) Message-Id: Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit Hi imagers, More precisely, I would like to eliminate the out of focus parts of my images. Than you. Bertrand ********************************************************** Bertrand MENARD University of Lausanne Université de Lausanne Institut of Ecology Institut d'écologie Plant Biology and Physiology Biologie et Physiologie Végétales Batiment de Biologie Batiment de Biologie CH-1015 LAUSANNE CH-1015 LAUSANNE Switzerland Suisse Tel : 00 41 21 692 42 19 Fax : 00 41 21 692 41 95 E-Mail : Bertrand.Menard@ie-bpv.unil.ch ********************************************************** ------------------------------ Date: Mon, 27 Sep 1999 15:45:36 +0200 From: "Peter van Loon" To: "NIH discussiegroep" Subject: What means tokenized? Message-ID: <022901bf08ee$a8c53b80$88bb7bd4@pcc42.pcc.agro.nl> Content-Type: multipart/alternative; boundary="----=_NextPart_000_0226_01BF08FF.578CEF40" This is a multi-part message in MIME format. ------=_NextPart_000_0226_01BF08FF.578CEF40 Content-Type: text/plain; charset="x-user-defined" Content-Transfer-Encoding: quoted-printable Hello, I make macro's in Object-Image. (Thank you Norbert Vischer for this = fantastic spin-off,=20 with the very usefull debugging tool). Unfortunately the used = text-editor in NIH and Object-Image=20 has a limitation of 32K (Apple's built in editor). So I have to be very = efficient whenever I write these macro's.=20 When I load the macro's, the info window gives information about = filesize and tells me that it is tokenized=20 (whatever it means?. It seems to be limited to 20K. Sometimes I reach = this limitation earlier then the=20 32K filesize limitation.=20 How can I prevent reaching this tokenized limit? Are there future plans of using a different editor without a 32K = limitation (whenever possible)? Regards, Peter van Loon Mushroom Experimental Station Horst, The Netherlands Mushpvl@wish.net=20 ------=_NextPart_000_0226_01BF08FF.578CEF40 Content-Type: text/html; charset="x-user-defined" Content-Transfer-Encoding: quoted-printable
Hello,
I make macro's in = Object-Image.=20 (Thank you Norbert Vischer for this fantastic spin-off,
with the = very=20 usefull debugging tool). Unfortunately the used text-editor in NIH and=20 Object-Image
has a limitation of 32K (Apple's built in editor). So I = have to=20 be very efficient whenever I write these macro's.
When I load the = macro's,=20 the info window gives information about filesize and tells me that it is = tokenized
(whatever it means?. It seems to be limited to 20K. = Sometimes I=20 reach this limitation earlier then the
32K filesize limitation. =
How can=20 I prevent reaching this tokenized limit?

Are there future plans of using a different editor without = a 32K=20 limitation (whenever possible)?
 
Regards,
 
Peter van = Loon
Mushroom Experimental=20 Station
Horst, The = Netherlands
Mushpvl@wish.net =
------=_NextPart_000_0226_01BF08FF.578CEF40-- ------------------------------ Date: Mon, 27 Sep 1999 17:10:32 +0200 From: Norbert Vischer To: nih-image@io.ece.drexel.edu Subject: Re: What means tokenized? Message-Id: Content-Type: text/plain; charset="us-ascii" > Hello, > I make macro's in Object-Image. (Thank you Norbert Vischer for this >fantastic spin-off, >with the very usefull debugging tool). Unfortunately the used text-editor >in NIH and Object-Image >has a limitation of 32K (Apple's built in editor). So I have to be very >efficient whenever I write these macro's. >When I load the macro's, the info window gives information about filesize >and tells me that it is tokenized >(whatever it means?. It seems to be limited to 20K. Sometimes I reach >this limitation earlier then the >32K filesize limitation. >How can I prevent reaching this tokenized limit? >Are there future plans of using a different editor without a 32K >limitation (whenever possible)? Regards, Peter van Loon Mushroom >Experimental Station Horst, The Netherlands Mushpvl@wish.net Peter: The macro text is dissolved into a stream of tokens, so it can be executed more efficiently. Usually, the token stream is shorter than the macro text, as variable names only occupy 2 bytes and comment is skipped. However, if you use a lot of string literals like s := 'This is a string literal', you can easily reach the 20k limit. Object-Image uses slightly more memory because of the extended set of macro commands. I can extend the stream size from 20 to 32K for you, but I don't have any plans to go beyond 32K. I'd rather implement a way to read/write from/to a generic file, so you can more easily interpret or manipulate files (text or other, any size) without using the Import-work-around. Here is an example how a macro is converted into a token stream: Macro text: =========== macro 'first macro'; var aa, bb: integer; begin aa := 2; bb := aa * 3; The above macro text fragment is converted into a token stream of 56 bytes: TokenStream =========== byte Token original text ---- ----- ------ 0 MacroT macro 'first macro'; 1 StringLiteral (11) 13 NullT 14 SemiColon 15 NewLineT 16 VarT var 17 NewLineT 18 Identifier aa, bb: integer; (2) 21 Comma 22 Identifier (2) 25 Colon 26 IntegerT 27 SemiColon 28 NewLineT 29 beginT begin 30 NewLineT 31 Identifier aa := 2; (2) 34 AssignOp 35 NumLiteral (4) 40 SemiColon 41 NewLine 42 Identifier bb := aa * 3; (2) 45 AssignOp 46 Identifier (2) 49 MulOp 50 NumericLiteral (4) 55 SemiColon 56 NewLine Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html -------------------------------- End of nih-image-d Digest V99 Issue #221 **************************************** From nih-image-request@io.ece.drexel.edu Tue Sep 28 07:57 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA07305; Tue, 28 Sep 1999 07:57:18 -0400 (EDT) Resent-Date: Tue, 28 Sep 1999 07:57:18 -0400 (EDT) Message-ID: <19990928114251.15995.qmail@hotmail.com> X-Originating-IP: [194.168.99.119] From: "Richard Herd" To: nih-image@io.ece.drexel.edu Subject: QuickTime Movie from NIH-Image Date: Tue, 28 Sep 1999 04:42:51 PDT Mime-Version: 1.0 Resent-Message-ID: <"679ZZ3.0.yD1.SdAyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1767 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; format=flowed Content-Length: 742 Status: O Hi I’d like to make a short QuickTime movie as a guide to using some User routines in NIH-Image. Is it possible to capture all screen activity for about 50 seconds or so, as a movie? The frame rate needn’t be high - probably 6 frames per second would be sufficient, but the frame size should probably not be below half the size of the screen being captured. I have a G3 PowerBook and an 8100/80AV available. Can a voiceover be added to the QuickTime movie afterwards? In theory it would save a lot of time typing in a manual, but not if I spend a couple of days puzzling it out. Thanks for any advice – Richard Herd ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Tue Sep 28 12:16 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA21723; Tue, 28 Sep 1999 12:16:45 -0400 (EDT) Resent-Date: Tue, 28 Sep 1999 12:16:45 -0400 (EDT) Message-Id: <37F0E569.56965918@purdue.edu> Date: Tue, 28 Sep 1999 10:57:29 -0500 From: Kris Jorgensen Organization: Purdue Univeristy X-Mailer: Mozilla 4.61 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: 32K file size limit in Object-Image References: Resent-Message-ID: <"lOM-W.0._F4.MJEyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1768 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: multipart/mixed; boundary="------------71001E321BE757E1CEA51627" Content-Length: 2269 Status: O This is a multi-part message in MIME format. --------------71001E321BE757E1CEA51627 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Norbert Vischer wrote: > > > Peter: > > The macro text is dissolved into a stream of tokens, so it can be executed > more efficiently. Usually, the token stream is shorter than the macro text, > as variable names only occupy 2 bytes and comment is skipped. However, if > you use a lot of string literals like s := 'This is a string literal', you > can easily reach the 20k limit. > Object-Image uses slightly more memory because of the extended set of macro > commands. > I can extend the stream size from 20 to 32K for you, but I don't have any > plans to go beyond 32K. I'd rather implement a way to read/write from/to a > generic file, so you can more easily interpret or manipulate files (text or > other, any size) without using the Import-work-around. > If the 32K limit is a result of using Apples built in text editor functions could you use WASTE which replaces the functions with one that no longer has the limit? Put my vote in for the "read/write from/to a generic file" capability!! I generate output from each slice in the stack since if the stack is large it overflows the 32K limit in the text window that I dump the results to. Now just write a file for each slice which can give me a lot of files to catanate together for further processing. What is the Import-work-around? Is it possible to load a macro on the fly? Thanks Kris -- Kris L. Jorgensen Office: ME 183 Phone: (765) 494-8585 Lab: (765) 484-8757 Email: kris@purdue.edu WWW: http://www.ecn.purdue.edu/~kris --------------71001E321BE757E1CEA51627 Content-Type: text/x-vcard; charset=us-ascii; name="kris.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Kris Jorgensen Content-Disposition: attachment; filename="kris.vcf" begin:vcard n:Jorgensen;Kris tel;fax:(765)494-0539 tel;work:(765)494-8585 x-mozilla-html:TRUE url:http://www.ecn.purdue.edu/~kris org:Purdue Univeristy;Mechanical Engineering adr:;;1288 Mechanical Engineering;West Lafayette;IN;47907; version:2.1 email;internet:kris@purdue.edu fn:Kris Jorgensen end:vcard --------------71001E321BE757E1CEA51627-- From nih-image-request@io.ece.drexel.edu Tue Sep 28 12:16 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA21761; Tue, 28 Sep 1999 12:16:53 -0400 (EDT) Resent-Date: Tue, 28 Sep 1999 12:16:53 -0400 (EDT) Message-ID: <19990928155747.65376.qmail@hotmail.com> X-Originating-IP: [129.22.235.1] From: "you se-ho" To: nih-image@io.ece.drexel.edu Cc: sxy28@po.cwru.edu Subject: [Resolution enhancement ?] Date: Tue, 28 Sep 1999 08:57:47 PDT Mime-Version: 1.0 Resent-Message-ID: <"M0uRX3.0.SP4.XMEyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1769 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; format=flowed Content-Length: 527 Status: O Hi, Imagers: I have Images which are obtained with microwave probe. Using the pointed image with very thin wire as a point spread function (PSF), I deconvoluted it from the measured images. Even though the resultant images look better than raw data (measured images), I doubt that the methods is called resolution enhancement. How about your opinion ? Any comments will be appreciated. Sincerely, SE-HO YOU ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Tue Sep 28 16:07 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA00744; Tue, 28 Sep 1999 16:07:27 -0400 (EDT) Resent-Date: Tue, 28 Sep 1999 16:07:27 -0400 (EDT) Message-Id: In-Reply-To: <37F0E569.56965918@purdue.edu> References: Mime-Version: 1.0 Date: Tue, 28 Sep 1999 21:50:36 +0200 To: nih-image@io.ece.drexel.edu From: Norbert Vischer Subject: Re: 32K file size limit in Object-Image Resent-Message-ID: <"iEAex.0.tx6.HpHyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1770 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 2742 Status: O My first priority is results window with no 32K limit, which will be finished soon. However, I don't have plans to substitute TextEdit by "Waste" to go beyond 32K in normal text windows. Also at the moment, I don't work on swappable macro text. Does your overflow occur due to writing results into a text window? Note that storing and retrieving results (real numbers) can be performed in Object-Image via user-created columns. This way you can accommodate much more numbers than would fit into a 32K window, and data are better organized. The columns are part of a double-clickable "object file" which also can include related auto-loadable/auto-running macro text, as well as non-destructive markers which appear as overlay in connected images. Columns can be either 'dynamic', which means that each entry is directly linked to a set of graphical objects called "cell", or they can be "static", where no link is maintained. As long as you perform data analysis within Object-Image, there is, due to the link, always a way to jump from the numbers (by double-clicking them) to the related images; e.g. you can detect suspect values and delete corresponding markers and numbers with the pistol tool. Here are a few related macro command descriptions as lised in the Help menu: MakeResultColumn('title') appends a dynamic resultcolumn. The title must be unique, otherwise a warning appears. InitColumn('title') Creates a new dynamic result column. If a column with this title already exists, it is cleared without warning and redefined as empty dynamic result column with no algorithm attached. InitStaticColumn('title') Creates a new static result column. If a column with this title already exists, it is cleared without warning and redefined as empty static result column. ClearColumn('ColumnName') Fills the column with NoValue DeleteColumn('ColumnName') Removes a column and its contents ColumnNumber('name'): integer If a result column or static column with this name is found, its number is returned, otherwise zero is returned. SetValue ('ColumnName', row, value) Writes a value into the result array. The column can be specified either by number or by the title string. GetValue ('ColumnName', row) : real Returns a value from the result array. The column can be specified either by number or by the title string. GetStatistics('column', 'string'): real returns a statistic value obtained from all non-empty values contained in a column, or returns NoValue if calculation is impossible. 'Column' must be the title one of the existing result columns (or alternatively the column number). 'String' must be one of: 'Count', 'Mean', 'Min', 'Max', 'StDev' or 'CV%'. From nih-image-request@io.ece.drexel.edu Tue Sep 28 17:21 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id RAA13525; Tue, 28 Sep 1999 17:21:13 -0400 (EDT) Resent-Date: Tue, 28 Sep 1999 17:21:13 -0400 (EDT) From: DrJohnRuss@aol.com Message-ID: Date: Tue, 28 Sep 1999 17:05:12 EDT Subject: Re: [Resolution enhancement ?] To: nih-image@io.ece.drexel.edu MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Mailer: AOL for Macintosh sub 56 Resent-Message-ID: <"W4d3E.0.bh2.xsIyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1771 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 716 Status: O In a message dated 9/28/99 12:10:40 PM, sxy28@hotmail.com writes: >I have Images which are obtained with microwave probe. Using the pointed >image with very thin wire as a point spread function (PSF), I deconvoluted >it from the measured images. Even though the resultant images look better >than raw data (measured images), I doubt that the methods is called >resolution enhancement. How about your opinion ? Sounds like a very clever way to measure the psf. In cases with low noise images, this is indeed a very effective way to improve image resolution, sometimes by as much as 40-50%. Is there a place where an example of your raw image, psf, and resultant image can be downloaded? I'd like to see it. From nih-image-request@io.ece.drexel.edu Tue Sep 28 18:11 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA22259; Tue, 28 Sep 1999 18:11:40 -0400 (EDT) Resent-Date: Tue, 28 Sep 1999 18:11:40 -0400 (EDT) Message-Id: <37F13B58.361AD5F0@purdue.edu> Date: Tue, 28 Sep 1999 17:04:14 -0500 From: Kris Jorgensen Organization: Purdue Univeristy X-Mailer: Mozilla 4.5 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: 32K file size limit in Object-Image References: Resent-Message-ID: <"ktCgU2.0.sr4.aeJyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1772 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: multipart/mixed; boundary="------------17273E9AA62F6B779254D6E6" Content-Length: 3581 Status: O This is a multi-part message in MIME format. --------------17273E9AA62F6B779254D6E6 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Norbert Vischer wrote: > > My first priority is results window with no 32K limit, which will be > finished soon. However, I don't have plans to substitute TextEdit by > "Waste" to go beyond 32K in normal text windows. Also at the moment, I > don't work on swappable macro text. > > Does your overflow occur due to writing results into a text window? Note > that storing and retrieving results (real numbers) can be performed in > Object-Image via user-created columns. This way you can accommodate much > more numbers than would fit into a 32K window, and data are better > organized. The columns are part of a double-clickable "object file" which > also can include related auto-loadable/auto-running macro text, as well as > non-destructive markers which appear as overlay in connected images. Yes, my over flow occurs when I'm writing results to a text window. I initially tried using this method but I had three problems. 1) When "Exporting Obj Results" the results listed in the file did not have enough resolution for small numbers. By dumping them to my own text window I can control the precision. 2) The other problem was that the columns are limited to 2500 rows. The data I generate comes from stacks that may have 10-200 slices. Each slice will have at least 36 and up to 5 times that many points (all in a single polygon cell) I need to output data for. This quickly fills up the 2500 rows. 3) Some of the columns data are intermediate values or the data applies to the cell and not the points in the cell and I don't want them in the output, but I get ALL of the data when exporting. So, if I could open a file (not a text window) for writing and just write to it and then close it I could accomplish what I would like to do. I don't think the over32K results window is going to help me here. I did once try to write the output to a image window but since I already have about 32K worth of macros I ran out of room and never finished trying this method. > > Columns can be either 'dynamic', which means that each entry is directly > linked to a set of graphical objects called "cell", or they can be > "static", where no link is maintained. As long as you perform data analysis > within Object-Image, there is, due to the link, always a way to jump from > the numbers (by double-clicking them) to the related images; e.g. you can > detect suspect values and delete corresponding markers and numbers with the > pistol tool. > These are wonderful extensions to Image and I can't say enough about how useful they are. BTW, is there a way to delete just a single point in a polygon cell? The pistol kills the entire cell and I only want a single point removed. -- Kris L. Jorgensen Office: ME 183 Phone: (765) 494-8585 Email: kris@purdue.edu WWW: http://www.ecn.purdue.edu/~kris --------------17273E9AA62F6B779254D6E6 Content-Type: text/x-vcard; charset=us-ascii; name="kris.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Kris Jorgensen Content-Disposition: attachment; filename="kris.vcf" begin:vcard n:Jorgensen;Kris tel;fax:(765)494-0539 tel;work:(765)494-8585 x-mozilla-html:TRUE url:http://www.ecn.purdue.edu/~kris org:Purdue Univeristy;Mechanical Engineering adr:;;1288 Mechanical Engineering;West Lafayette;IN;47907; version:2.1 email;internet:kris@purdue.edu fn:Kris Jorgensen end:vcard --------------17273E9AA62F6B779254D6E6-- From nih-image-d-request@io.ece.drexel.edu Wed Sep 29 06:16 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA20862; Wed, 29 Sep 1999 06:16:08 -0400 (EDT) Date: Wed, 29 Sep 1999 06:16:08 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199909291016.GAA20862@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #222 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/222 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 13023 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 222 Today's Topics: QuickTime Movie from NIH-Image [ "Richard Herd" ] 32K file size limit in Object-Image [ Kris Jorgensen ] [Resolution enhancement ?] [ "you se-ho" ] Re: 32K file size limit in Object-Im [ Norbert Vischer ] ------------------------------ Date: Tue, 28 Sep 1999 04:42:51 PDT From: "Richard Herd" To: nih-image@io.ece.drexel.edu Subject: QuickTime Movie from NIH-Image Message-ID: <19990928114251.15995.qmail@hotmail.com> Content-Type: text/plain; format=flowed Hi I’d like to make a short QuickTime movie as a guide to using some User routines in NIH-Image. Is it possible to capture all screen activity for about 50 seconds or so, as a movie? The frame rate needn’t be high - probably 6 frames per second would be sufficient, but the frame size should probably not be below half the size of the screen being captured. I have a G3 PowerBook and an 8100/80AV available. Can a voiceover be added to the QuickTime movie afterwards? In theory it would save a lot of time typing in a manual, but not if I spend a couple of days puzzling it out. Thanks for any advice – Richard Herd ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com ------------------------------ Date: Tue, 28 Sep 1999 10:57:29 -0500 From: Kris Jorgensen To: nih-image@io.ece.drexel.edu Subject: 32K file size limit in Object-Image Message-Id: <37F0E569.56965918@purdue.edu> Content-Type: multipart/mixed; boundary="------------71001E321BE757E1CEA51627" This is a multi-part message in MIME format. --------------71001E321BE757E1CEA51627 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Norbert Vischer wrote: > > > Peter: > > The macro text is dissolved into a stream of tokens, so it can be executed > more efficiently. Usually, the token stream is shorter than the macro text, > as variable names only occupy 2 bytes and comment is skipped. However, if > you use a lot of string literals like s := 'This is a string literal', you > can easily reach the 20k limit. > Object-Image uses slightly more memory because of the extended set of macro > commands. > I can extend the stream size from 20 to 32K for you, but I don't have any > plans to go beyond 32K. I'd rather implement a way to read/write from/to a > generic file, so you can more easily interpret or manipulate files (text or > other, any size) without using the Import-work-around. > If the 32K limit is a result of using Apples built in text editor functions could you use WASTE which replaces the functions with one that no longer has the limit? Put my vote in for the "read/write from/to a generic file" capability!! I generate output from each slice in the stack since if the stack is large it overflows the 32K limit in the text window that I dump the results to. Now just write a file for each slice which can give me a lot of files to catanate together for further processing. What is the Import-work-around? Is it possible to load a macro on the fly? Thanks Kris -- Kris L. Jorgensen Office: ME 183 Phone: (765) 494-8585 Lab: (765) 484-8757 Email: kris@purdue.edu WWW: http://www.ecn.purdue.edu/~kris --------------71001E321BE757E1CEA51627 Content-Type: text/x-vcard; charset=us-ascii; name="kris.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Kris Jorgensen Content-Disposition: attachment; filename="kris.vcf" begin:vcard n:Jorgensen;Kris tel;fax:(765)494-0539 tel;work:(765)494-8585 x-mozilla-html:TRUE url:http://www.ecn.purdue.edu/~kris org:Purdue Univeristy;Mechanical Engineering adr:;;1288 Mechanical Engineering;West Lafayette;IN;47907; version:2.1 email;internet:kris@purdue.edu fn:Kris Jorgensen end:vcard --------------71001E321BE757E1CEA51627-- ------------------------------ Date: Tue, 28 Sep 1999 08:57:47 PDT From: "you se-ho" To: nih-image@io.ece.drexel.edu Cc: sxy28@po.cwru.edu Subject: [Resolution enhancement ?] Message-ID: <19990928155747.65376.qmail@hotmail.com> Content-Type: text/plain; format=flowed Hi, Imagers: I have Images which are obtained with microwave probe. Using the pointed image with very thin wire as a point spread function (PSF), I deconvoluted it from the measured images. Even though the resultant images look better than raw data (measured images), I doubt that the methods is called resolution enhancement. How about your opinion ? Any comments will be appreciated. Sincerely, SE-HO YOU ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com ------------------------------ Date: Tue, 28 Sep 1999 21:50:36 +0200 From: Norbert Vischer To: nih-image@io.ece.drexel.edu Subject: Re: 32K file size limit in Object-Image Message-Id: Content-Type: text/plain; charset="us-ascii" My first priority is results window with no 32K limit, which will be finished soon. However, I don't have plans to substitute TextEdit by "Waste" to go beyond 32K in normal text windows. Also at the moment, I don't work on swappable macro text. Does your overflow occur due to writing results into a text window? Note that storing and retrieving results (real numbers) can be performed in Object-Image via user-created columns. This way you can accommodate much more numbers than would fit into a 32K window, and data are better organized. The columns are part of a double-clickable "object file" which also can include related auto-loadable/auto-running macro text, as well as non-destructive markers which appear as overlay in connected images. Columns can be either 'dynamic', which means that each entry is directly linked to a set of graphical objects called "cell", or they can be "static", where no link is maintained. As long as you perform data analysis within Object-Image, there is, due to the link, always a way to jump from the numbers (by double-clicking them) to the related images; e.g. you can detect suspect values and delete corresponding markers and numbers with the pistol tool. Here are a few related macro command descriptions as lised in the Help menu: MakeResultColumn('title') appends a dynamic resultcolumn. The title must be unique, otherwise a warning appears. InitColumn('title') Creates a new dynamic result column. If a column with this title already exists, it is cleared without warning and redefined as empty dynamic result column with no algorithm attached. InitStaticColumn('title') Creates a new static result column. If a column with this title already exists, it is cleared without warning and redefined as empty static result column. ClearColumn('ColumnName') Fills the column with NoValue DeleteColumn('ColumnName') Removes a column and its contents ColumnNumber('name'): integer If a result column or static column with this name is found, its number is returned, otherwise zero is returned. SetValue ('ColumnName', row, value) Writes a value into the result array. The column can be specified either by number or by the title string. GetValue ('ColumnName', row) : real Returns a value from the result array. The column can be specified either by number or by the title string. GetStatistics('column', 'string'): real returns a statistic value obtained from all non-empty values contained in a column, or returns NoValue if calculation is impossible. 'Column' must be the title one of the existing result columns (or alternatively the column number). 'String' must be one of: 'Count', 'Mean', 'Min', 'Max', 'StDev' or 'CV%'. ------------------------------ Date: Tue, 28 Sep 1999 17:05:12 EDT From: DrJohnRuss@aol.com To: nih-image@io.ece.drexel.edu Subject: Re: [Resolution enhancement ?] Message-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit In a message dated 9/28/99 12:10:40 PM, sxy28@hotmail.com writes: >I have Images which are obtained with microwave probe. Using the pointed >image with very thin wire as a point spread function (PSF), I deconvoluted >it from the measured images. Even though the resultant images look better >than raw data (measured images), I doubt that the methods is called >resolution enhancement. How about your opinion ? Sounds like a very clever way to measure the psf. In cases with low noise images, this is indeed a very effective way to improve image resolution, sometimes by as much as 40-50%. Is there a place where an example of your raw image, psf, and resultant image can be downloaded? I'd like to see it. ------------------------------ Date: Tue, 28 Sep 1999 17:04:14 -0500 From: Kris Jorgensen To: nih-image@io.ece.drexel.edu Subject: Re: 32K file size limit in Object-Image Message-Id: <37F13B58.361AD5F0@purdue.edu> Content-Type: multipart/mixed; boundary="------------17273E9AA62F6B779254D6E6" This is a multi-part message in MIME format. --------------17273E9AA62F6B779254D6E6 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Norbert Vischer wrote: > > My first priority is results window with no 32K limit, which will be > finished soon. However, I don't have plans to substitute TextEdit by > "Waste" to go beyond 32K in normal text windows. Also at the moment, I > don't work on swappable macro text. > > Does your overflow occur due to writing results into a text window? Note > that storing and retrieving results (real numbers) can be performed in > Object-Image via user-created columns. This way you can accommodate much > more numbers than would fit into a 32K window, and data are better > organized. The columns are part of a double-clickable "object file" which > also can include related auto-loadable/auto-running macro text, as well as > non-destructive markers which appear as overlay in connected images. Yes, my over flow occurs when I'm writing results to a text window. I initially tried using this method but I had three problems. 1) When "Exporting Obj Results" the results listed in the file did not have enough resolution for small numbers. By dumping them to my own text window I can control the precision. 2) The other problem was that the columns are limited to 2500 rows. The data I generate comes from stacks that may have 10-200 slices. Each slice will have at least 36 and up to 5 times that many points (all in a single polygon cell) I need to output data for. This quickly fills up the 2500 rows. 3) Some of the columns data are intermediate values or the data applies to the cell and not the points in the cell and I don't want them in the output, but I get ALL of the data when exporting. So, if I could open a file (not a text window) for writing and just write to it and then close it I could accomplish what I would like to do. I don't think the over32K results window is going to help me here. I did once try to write the output to a image window but since I already have about 32K worth of macros I ran out of room and never finished trying this method. > > Columns can be either 'dynamic', which means that each entry is directly > linked to a set of graphical objects called "cell", or they can be > "static", where no link is maintained. As long as you perform data analysis > within Object-Image, there is, due to the link, always a way to jump from > the numbers (by double-clicking them) to the related images; e.g. you can > detect suspect values and delete corresponding markers and numbers with the > pistol tool. > These are wonderful extensions to Image and I can't say enough about how useful they are. BTW, is there a way to delete just a single point in a polygon cell? The pistol kills the entire cell and I only want a single point removed. -- Kris L. Jorgensen Office: ME 183 Phone: (765) 494-8585 Email: kris@purdue.edu WWW: http://www.ecn.purdue.edu/~kris --------------17273E9AA62F6B779254D6E6 Content-Type: text/x-vcard; charset=us-ascii; name="kris.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Kris Jorgensen Content-Disposition: attachment; filename="kris.vcf" begin:vcard n:Jorgensen;Kris tel;fax:(765)494-0539 tel;work:(765)494-8585 x-mozilla-html:TRUE url:http://www.ecn.purdue.edu/~kris org:Purdue Univeristy;Mechanical Engineering adr:;;1288 Mechanical Engineering;West Lafayette;IN;47907; version:2.1 email;internet:kris@purdue.edu fn:Kris Jorgensen end:vcard --------------17273E9AA62F6B779254D6E6-- -------------------------------- End of nih-image-d Digest V99 Issue #222 **************************************** From nih-image-request@io.ece.drexel.edu Wed Sep 29 06:59 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA28334; Wed, 29 Sep 1999 06:59:48 -0400 (EDT) Resent-Date: Wed, 29 Sep 1999 06:59:48 -0400 (EDT) X-Authentication-Warning: mail.bio.uva.nl: Host gold.bio.uva.nl [145.18.160.48] claimed to be [145.18.160.48] Message-Id: In-Reply-To: <37F13B58.361AD5F0@purdue.edu> References: Mime-Version: 1.0 Date: Wed, 29 Sep 1999 12:48:02 +0200 To: nih-image@io.ece.drexel.edu From: Norbert Vischer Subject: Re: 32K file size limit in Object-Image Resent-Message-ID: <"2XH7u1.0.e46.XnUyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1773 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1608 Status: O 1) Precision of "Export Object results" can be controlled by "Analyze Menu: Options". (Sorry, this was only mentioned in the help balloons, not in the manual). 2) I am busy with a new design of the results, with less limitations for cell count, and including a way to mask and selectively export columns. Until then, one has to do it in a spreadsheet program. 3) Another way to export data is via menu "Export XYZ data", which exports all point positions of objects, but not the result columns. Although this file is saved as a Rotater file, it is a normal text file which can be read by a spreadsheet program or text processor. Like other export functions, this output is not limited to 32K and is structured in a way that the data organisation is not lost. Precision is controllable as mentioned above. As stated before, the results window is my first priority (before generic file access). You can re-open a cell by double-clicking it with the Object-Select (Finger) tool (the corresponding macro command is OpenCell(n) - not yet documented). Once a cell is open, it appears with an italic number label, and you can delete points sequentially from back to front with the backspace key. Deleting a random point in a polygon is not yet supported - you only can drag it to a different position. Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html From nih-image-request@io.ece.drexel.edu Thu Sep 30 05:25 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA04131; Thu, 30 Sep 1999 05:25:57 -0400 (EDT) Resent-Date: Thu, 30 Sep 1999 05:25:57 -0400 (EDT) Message-ID: From: Gary Chinga To: "'nih-image@io.ece.drexel.edu'" Subject: Histogram. Date: Thu, 30 Sep 1999 11:11:35 +0200 X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.996.62 MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Resent-Message-ID: <"D6eKl2.0.MG.mYoyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1774 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 320 Status: O I have a histogram of an image showing three tops and I want to get the pixelvalue of the minimal point between the second and third top. How can I do that? I have the image and the histogram in the following page if someone didnt understand my explanation. http://home.nvg.ntnu.no/~gary/Temp/histogram.gif >Gary. From nih-image-d-request@io.ece.drexel.edu Thu Sep 30 05:28 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA04533; Thu, 30 Sep 1999 05:28:04 -0400 (EDT) Date: Thu, 30 Sep 1999 05:28:04 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199909300928.FAA04533@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #223 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/223 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 2953 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 223 Today's Topics: Re: 32K file size limit in Object-Im [ Norbert Vischer ] ------------------------------ Date: Wed, 29 Sep 1999 12:48:02 +0200 From: Norbert Vischer To: nih-image@io.ece.drexel.edu Subject: Re: 32K file size limit in Object-Image Message-Id: Content-Type: text/plain; charset="us-ascii" 1) Precision of "Export Object results" can be controlled by "Analyze Menu: Options". (Sorry, this was only mentioned in the help balloons, not in the manual). 2) I am busy with a new design of the results, with less limitations for cell count, and including a way to mask and selectively export columns. Until then, one has to do it in a spreadsheet program. 3) Another way to export data is via menu "Export XYZ data", which exports all point positions of objects, but not the result columns. Although this file is saved as a Rotater file, it is a normal text file which can be read by a spreadsheet program or text processor. Like other export functions, this output is not limited to 32K and is structured in a way that the data organisation is not lost. Precision is controllable as mentioned above. As stated before, the results window is my first priority (before generic file access). You can re-open a cell by double-clicking it with the Object-Select (Finger) tool (the corresponding macro command is OpenCell(n) - not yet documented). Once a cell is open, it appears with an italic number label, and you can delete points sequentially from back to front with the backspace key. Deleting a random point in a polygon is not yet supported - you only can drag it to a different position. Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html ------------------------------ Date: Thu, 30 Sep 1999 11:11:35 +0200 From: Gary Chinga To: "'nih-image@io.ece.drexel.edu'" Subject: Histogram. Message-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit I have a histogram of an image showing three tops and I want to get the pixelvalue of the minimal point between the second and third top. How can I do that? I have the image and the histogram in the following page if someone didnt understand my explanation. http://home.nvg.ntnu.no/~gary/Temp/histogram.gif >Gary. -------------------------------- End of nih-image-d Digest V99 Issue #223 **************************************** From nih-image-request@io.ece.drexel.edu Thu Sep 30 05:43 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA07351; Thu, 30 Sep 1999 05:43:34 -0400 (EDT) Resent-Date: Thu, 30 Sep 1999 05:43:34 -0400 (EDT) Message-ID: <003e01bf0b26$462c14a0$28807ac1@seer.aii.co.uk> From: "Shiddhartha Nandy" To: "nih listserver - posting" Subject: Re: [Resolution enhancement ?] Date: Thu, 30 Sep 1999 10:29:20 +0100 MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.72.3115.0 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 X-MDaemon-Deliver-To: nih-image@biomed.drexel.edu X-Return-Path: sn@aii.co.uk Resent-Message-ID: <"NEraC2.0.f31.gpoyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1775 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="iso-8859-1" Content-Length: 914 Status: O I'd be interested in seeing the images too. Dr. Shiddhartha Nandy Senior Software Engineer Applied Imaging International Ltd. Tel: +44 (0) 191 202 3100 Fax: +44 (0) 191 202 3101 sn@aii.co.uk >In a message dated 9/28/99 12:10:40 PM, sxy28@hotmail.com writes: >>I have Images which are obtained with microwave probe. Using the pointed >>image with very thin wire as a point spread function (PSF), I deconvoluted >>it from the measured images. Even though the resultant images look better >>than raw data (measured images), I doubt that the methods is called >>resolution enhancement. How about your opinion ? >Sounds like a very clever way to measure the psf. In cases with low noise >images, this is indeed a very effective way to improve image resolution, >sometimes by as much as 40-50%. Is there a place where an example of your raw >image, psf, and resultant image can be downloaded? I'd like to see it. From nih-image-request@io.ece.drexel.edu Thu Sep 30 08:47 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA05068; Thu, 30 Sep 1999 08:47:21 -0400 (EDT) Resent-Date: Thu, 30 Sep 1999 08:47:21 -0400 (EDT) Message-Id: <199909301231.IAA02400@io.ece.drexel.edu> Subject: Blank fields and saved images Date: Thu, 30 Sep 99 07:32:36 -0500 x-sender: rblyston@exchange.trinity.edu x-mailer: Claris Emailer 1.1 From: To: "NIH-Image List" Mime-Version: 1.0 Resent-Message-ID: <"za8nj.0.gb.WWryt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1776 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 821 Status: O To the List: I trust I have a mundane question. In Image one can save a blank field and have it subtracted from the live image: Nice when one has illumination or dirt problems on microscope lenses. Just how does Image do this? I have saved captured images for which I need to go back and subtract a blank field; however, when I do so, various default settings come on board or if I zero them, the image is changed far more than when the live subtraction is performed. I would appreciate info on what settings and how the blank field subtraction is performed on a live image so that I could duplicate the procedure on saved images. Thanks. Blystone in Texas Robert V. Blystone, Ph.D. Professor of Biology Trinity University San Antonio, Texas 78212 210.999-7243 210.999-7229 FAX From nih-image-request@io.ece.drexel.edu Thu Sep 30 11:38 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA06218; Thu, 30 Sep 1999 11:38:08 -0400 (EDT) Resent-Date: Thu, 30 Sep 1999 11:38:08 -0400 (EDT) Subject: Re: Blank fields and saved images Date: Thu, 30 Sep 99 11:24:18 -0400 x-sender: jlr$biol@qc1.qc.edu x-mailer: Claris Emailer 1.1 From: jared rifkin To: Mime-Version: 1.0 Message-ID: <4610AD743C@qc1.qc.edu> Resent-Message-ID: <"GG9AK1.0.du.l2uyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1777 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 55 Status: O please post answers to this question to entire group. From nih-image-request@io.ece.drexel.edu Thu Sep 30 13:29 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA24543; Thu, 30 Sep 1999 13:29:17 -0400 (EDT) Resent-Date: Thu, 30 Sep 1999 13:29:17 -0400 (EDT) Message-ID: <19990930171207.39212.qmail@hotmail.com> X-Originating-IP: [129.22.56.251] From: "you se-ho" To: nih-image@io.ece.drexel.edu Subject: Re: [Resolution enhancement ?] Date: Thu, 30 Sep 1999 10:12:07 PDT Mime-Version: 1.0 Resent-Message-ID: <"PwzJI.0.PI5.Eevyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1778 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; format=flowed Content-Length: 2433 Status: O Hi, Imagers I have no way to post the images. However, I am trying to think about it. Thank you. Sincerely, SE-HO YOU >From: "Shiddhartha Nandy" >Reply-To: nih-image@io.ece.drexel.edu >To: "nih listserver - posting" >Subject: Re: [Resolution enhancement ?] >Date: Thu, 30 Sep 1999 10:29:20 +0100 >MIME-Version: 1.0 >Return-Path: nih-image-request@io.ece.drexel.edu >Received: from io.ece.drexel.edu (lists@io.ece.drexel.edu [144.118.32.3]) >by arthur.INS.CWRU.Edu with ESMTP (8.8.8+cwru/CWRU-3.6)id FAA08703; Thu, 30 >Sep 1999 05:46:29 -0400 (EDT) (from nih-image-request@io.ece.drexel.edu) >Received: (from lists@localhost)by io.ece.drexel.edu (8.8.8/8.8.8) id >FAA07370;Thu, 30 Sep 1999 05:43:38 -0400 (EDT) >Resent-Date: Thu, 30 Sep 1999 05:43:38 -0400 (EDT) >Message-ID: <003e01bf0b26$462c14a0$28807ac1@seer.aii.co.uk> >X-Priority: 3 >X-MSMail-Priority: Normal >X-Mailer: Microsoft Outlook Express 4.72.3115.0 >X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 >X-MDaemon-Deliver-To: nih-image@biomed.drexel.edu >X-Return-Path: sn@aii.co.uk >Resent-Message-ID: <"NEraC2.0.f31.gpoyt"@io> >Resent-From: nih-image@io.ece.drexel.edu >X-Mailing-List: archive/latest/1775 >X-Loop: nih-image@biomed.drexel.edu >Precedence: list >Resent-Sender: nih-image-request@io.ece.drexel.edu > >I'd be interested in seeing the images too. > >Dr. Shiddhartha Nandy >Senior Software Engineer >Applied Imaging International Ltd. >Tel: +44 (0) 191 202 3100 >Fax: +44 (0) 191 202 3101 >sn@aii.co.uk > > >In a message dated 9/28/99 12:10:40 PM, sxy28@hotmail.com writes: > > >>I have Images which are obtained with microwave probe. Using the pointed > >>image with very thin wire as a point spread function (PSF), I >deconvoluted > >>it from the measured images. Even though the resultant images look >better > >>than raw data (measured images), I doubt that the methods is called > >>resolution enhancement. How about your opinion ? > > >Sounds like a very clever way to measure the psf. In cases with low noise > >images, this is indeed a very effective way to improve image resolution, > >sometimes by as much as 40-50%. Is there a place where an example of your >raw > >image, psf, and resultant image can be downloaded? I'd like to see it. > > > > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Thu Sep 30 14:15 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id OAA04657; Thu, 30 Sep 1999 14:15:44 -0400 (EDT) Resent-Date: Thu, 30 Sep 1999 14:15:44 -0400 (EDT) Date: Thu, 30 Sep 1999 13:40:48 -0400 From: "Barry R. Bickmore" Subject: Re: [Resolution enhancement ?] In-reply-to: <19990930171207.39212.qmail@hotmail.com> X-Sender: bbickmor@mail.vt.edu To: nih-image@io.ece.drexel.edu Message-id: MIME-version: 1.0 Content-transfer-encoding: 8BIT Resent-Message-ID: <"1Cs-62.0.XF7.sIwyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1779 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 778 Status: O >Hi, Imagers > >I have no way to post the images. However, I am trying to think about it. Try getting a free web site from http://www.geocities.com . They will give you 11 Mb of space. Barry _____________________________________________________________________________ Barry Bickmore "Lisa, just because Dept. of Geological Sciences I don't care doesn't mean Virginia Tech I don't understand!" Blacksburg, VA 24061 -Homer J. Simpson bbickmor@vt.edu (540)231-8575 FAX:(540)231-3386 Barry's preprints/reprints page: http://www.geocities.com/Athens/Parthenon/2671/preprint.html _____________________________________________________________________________ť From nih-image-request@io.ece.drexel.edu Thu Sep 30 16:13 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA22550; Thu, 30 Sep 1999 16:13:05 -0400 (EDT) Resent-Date: Thu, 30 Sep 1999 16:13:05 -0400 (EDT) Message-ID: From: "Heeschen, Bill (WA)" To: "'nih-image@io.ece.drexel.edu'" Subject: RE: Blank fields and saved images Date: Thu, 30 Sep 1999 14:05:48 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2448.0) Resent-Message-ID: <"Ax-La1.0.rv4.G3yyt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1780 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="iso-8859-1" Content-Length: 2665 Status: O Robert/all I have gone into the source code and figured it out and have implemented a couple of macros to accomplish essentially the same thing. Note that these macros are from a much larger file which contains a number of global variables. If you can't see it declared locally in the macro, it's done as a global. These two variables were set up as global variables with default values: mathScale:= 1; {default} mathOffset:=65; {default value, automatically set to mean of background} currSamp is a global string that contains the name of the current sample being analyzed. I use the following macro to "set up" the current image as the background image macro 'Set Background [S]'; var n,mean,mode,min,max:integer; begin if not(initDone) then Initialize; bkgndName:= WindowTitle; SelectAll; Measure; KillROI; GetResults(n,mean,mode,min,max); mathOffset:=mean; BkgExists:= true; bkgndPID:= PidNumber; BkgCheck:= true; ShowMessage('Do not close background window!'); end; this is the macro code that does the current Background from the current Image and creates a new image with the prefix "B" to indicate Background correction: macro 'Subtract Current Background [B]' var rawSamplePID: integer; begin currSamp:= WindowTitle; currSampPID:= PidNumber; if BkgExists and PidExists(bkgndPID) then begin currSamp:= concat('B',currSamp); Duplicate('Raw_Sample'); rawSamplePID:= PidNumber; ImageMath('sub',rawSamplePID,bkgndPID,mathScale,mathOffset,currSamp); currSampPID:= PidNumber; SelectPic(rawSamplePID); Dispose; SelectPic(currSampPID); end else PutMessage('No current background image!'); end; -----Original Message----- From: rblyston@trinity.edu [mailto:rblyston@trinity.edu] Sent: Thursday, September 30, 1999 8:33 AM To: NIH-Image List Subject: Blank fields and saved images To the List: I trust I have a mundane question. In Image one can save a blank field and have it subtracted from the live image: Nice when one has illumination or dirt problems on microscope lenses. Just how does Image do this? I have saved captured images for which I need to go back and subtract a blank field; however, when I do so, various default settings come on board or if I zero them, the image is changed far more than when the live subtraction is performed. I would appreciate info on what settings and how the blank field subtraction is performed on a live image so that I could duplicate the procedure on saved images. Thanks. Blystone in Texas Robert V. Blystone, Ph.D. Professor of Biology Trinity University San Antonio, Texas 78212 210.999-7243 210.999-7229 FAX From nih-image-d-request@io.ece.drexel.edu Fri Oct 1 06:23 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA23005; Fri, 1 Oct 1999 06:23:14 -0400 (EDT) Date: Fri, 1 Oct 1999 06:23:14 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910011023.GAA23005@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #224 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/224 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 10207 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 224 Today's Topics: Re: [Resolution enhancement ?] [ "Shiddhartha Nandy" ] Blank fields and saved images [ ] Re: Blank fields and saved images [ jared rifkin ] Re: [Resolution enhancement ?] [ "you se-ho" ] Re: [Resolution enhancement ?] [ "Barry R. Bickmore" To: "nih listserver - posting" Subject: Re: [Resolution enhancement ?] Message-ID: <003e01bf0b26$462c14a0$28807ac1@seer.aii.co.uk> Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit I'd be interested in seeing the images too. Dr. Shiddhartha Nandy Senior Software Engineer Applied Imaging International Ltd. Tel: +44 (0) 191 202 3100 Fax: +44 (0) 191 202 3101 sn@aii.co.uk >In a message dated 9/28/99 12:10:40 PM, sxy28@hotmail.com writes: >>I have Images which are obtained with microwave probe. Using the pointed >>image with very thin wire as a point spread function (PSF), I deconvoluted >>it from the measured images. Even though the resultant images look better >>than raw data (measured images), I doubt that the methods is called >>resolution enhancement. How about your opinion ? >Sounds like a very clever way to measure the psf. In cases with low noise >images, this is indeed a very effective way to improve image resolution, >sometimes by as much as 40-50%. Is there a place where an example of your raw >image, psf, and resultant image can be downloaded? I'd like to see it. ------------------------------ Date: Thu, 30 Sep 99 07:32:36 -0500 From: To: "NIH-Image List" Subject: Blank fields and saved images Message-Id: <199909301231.IAA02400@io.ece.drexel.edu> Content-Type: text/plain; charset="US-ASCII" To the List: I trust I have a mundane question. In Image one can save a blank field and have it subtracted from the live image: Nice when one has illumination or dirt problems on microscope lenses. Just how does Image do this? I have saved captured images for which I need to go back and subtract a blank field; however, when I do so, various default settings come on board or if I zero them, the image is changed far more than when the live subtraction is performed. I would appreciate info on what settings and how the blank field subtraction is performed on a live image so that I could duplicate the procedure on saved images. Thanks. Blystone in Texas Robert V. Blystone, Ph.D. Professor of Biology Trinity University San Antonio, Texas 78212 210.999-7243 210.999-7229 FAX ------------------------------ Date: Thu, 30 Sep 99 11:24:18 -0400 From: jared rifkin To: Subject: Re: Blank fields and saved images Message-ID: <4610AD743C@qc1.qc.edu> Content-Type: text/plain; charset="US-ASCII" please post answers to this question to entire group. ------------------------------ Date: Thu, 30 Sep 1999 10:12:07 PDT From: "you se-ho" To: nih-image@io.ece.drexel.edu Subject: Re: [Resolution enhancement ?] Message-ID: <19990930171207.39212.qmail@hotmail.com> Content-Type: text/plain; format=flowed Hi, Imagers I have no way to post the images. However, I am trying to think about it. Thank you. Sincerely, SE-HO YOU >From: "Shiddhartha Nandy" >Reply-To: nih-image@io.ece.drexel.edu >To: "nih listserver - posting" >Subject: Re: [Resolution enhancement ?] >Date: Thu, 30 Sep 1999 10:29:20 +0100 >MIME-Version: 1.0 >Return-Path: nih-image-request@io.ece.drexel.edu >Received: from io.ece.drexel.edu (lists@io.ece.drexel.edu [144.118.32.3]) >by arthur.INS.CWRU.Edu with ESMTP (8.8.8+cwru/CWRU-3.6)id FAA08703; Thu, 30 >Sep 1999 05:46:29 -0400 (EDT) (from nih-image-request@io.ece.drexel.edu) >Received: (from lists@localhost)by io.ece.drexel.edu (8.8.8/8.8.8) id >FAA07370;Thu, 30 Sep 1999 05:43:38 -0400 (EDT) >Resent-Date: Thu, 30 Sep 1999 05:43:38 -0400 (EDT) >Message-ID: <003e01bf0b26$462c14a0$28807ac1@seer.aii.co.uk> >X-Priority: 3 >X-MSMail-Priority: Normal >X-Mailer: Microsoft Outlook Express 4.72.3115.0 >X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3110.3 >X-MDaemon-Deliver-To: nih-image@biomed.drexel.edu >X-Return-Path: sn@aii.co.uk >Resent-Message-ID: <"NEraC2.0.f31.gpoyt"@io> >Resent-From: nih-image@io.ece.drexel.edu >X-Mailing-List: archive/latest/1775 >X-Loop: nih-image@biomed.drexel.edu >Precedence: list >Resent-Sender: nih-image-request@io.ece.drexel.edu > >I'd be interested in seeing the images too. > >Dr. Shiddhartha Nandy >Senior Software Engineer >Applied Imaging International Ltd. >Tel: +44 (0) 191 202 3100 >Fax: +44 (0) 191 202 3101 >sn@aii.co.uk > > >In a message dated 9/28/99 12:10:40 PM, sxy28@hotmail.com writes: > > >>I have Images which are obtained with microwave probe. Using the pointed > >>image with very thin wire as a point spread function (PSF), I >deconvoluted > >>it from the measured images. Even though the resultant images look >better > >>than raw data (measured images), I doubt that the methods is called > >>resolution enhancement. How about your opinion ? > > >Sounds like a very clever way to measure the psf. In cases with low noise > >images, this is indeed a very effective way to improve image resolution, > >sometimes by as much as 40-50%. Is there a place where an example of your >raw > >image, psf, and resultant image can be downloaded? I'd like to see it. > > > > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com ------------------------------ Date: Thu, 30 Sep 1999 13:40:48 -0400 From: "Barry R. Bickmore" To: nih-image@io.ece.drexel.edu Subject: Re: [Resolution enhancement ?] Message-id: Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 8BIT >Hi, Imagers > >I have no way to post the images. However, I am trying to think about it. Try getting a free web site from http://www.geocities.com . They will give you 11 Mb of space. Barry _____________________________________________________________________________ Barry Bickmore "Lisa, just because Dept. of Geological Sciences I don't care doesn't mean Virginia Tech I don't understand!" Blacksburg, VA 24061 -Homer J. Simpson bbickmor@vt.edu (540)231-8575 FAX:(540)231-3386 Barry's preprints/reprints page: http://www.geocities.com/Athens/Parthenon/2671/preprint.html _____________________________________________________________________________ť ------------------------------ Date: Thu, 30 Sep 1999 14:05:48 -0400 From: "Heeschen, Bill (WA)" To: "'nih-image@io.ece.drexel.edu'" Subject: RE: Blank fields and saved images Message-ID: Content-Type: text/plain; charset="iso-8859-1" Robert/all I have gone into the source code and figured it out and have implemented a couple of macros to accomplish essentially the same thing. Note that these macros are from a much larger file which contains a number of global variables. If you can't see it declared locally in the macro, it's done as a global. These two variables were set up as global variables with default values: mathScale:= 1; {default} mathOffset:=65; {default value, automatically set to mean of background} currSamp is a global string that contains the name of the current sample being analyzed. I use the following macro to "set up" the current image as the background image macro 'Set Background [S]'; var n,mean,mode,min,max:integer; begin if not(initDone) then Initialize; bkgndName:= WindowTitle; SelectAll; Measure; KillROI; GetResults(n,mean,mode,min,max); mathOffset:=mean; BkgExists:= true; bkgndPID:= PidNumber; BkgCheck:= true; ShowMessage('Do not close background window!'); end; this is the macro code that does the current Background from the current Image and creates a new image with the prefix "B" to indicate Background correction: macro 'Subtract Current Background [B]' var rawSamplePID: integer; begin currSamp:= WindowTitle; currSampPID:= PidNumber; if BkgExists and PidExists(bkgndPID) then begin currSamp:= concat('B',currSamp); Duplicate('Raw_Sample'); rawSamplePID:= PidNumber; ImageMath('sub',rawSamplePID,bkgndPID,mathScale,mathOffset,currSamp); currSampPID:= PidNumber; SelectPic(rawSamplePID); Dispose; SelectPic(currSampPID); end else PutMessage('No current background image!'); end; -----Original Message----- From: rblyston@trinity.edu [mailto:rblyston@trinity.edu] Sent: Thursday, September 30, 1999 8:33 AM To: NIH-Image List Subject: Blank fields and saved images To the List: I trust I have a mundane question. In Image one can save a blank field and have it subtracted from the live image: Nice when one has illumination or dirt problems on microscope lenses. Just how does Image do this? I have saved captured images for which I need to go back and subtract a blank field; however, when I do so, various default settings come on board or if I zero them, the image is changed far more than when the live subtraction is performed. I would appreciate info on what settings and how the blank field subtraction is performed on a live image so that I could duplicate the procedure on saved images. Thanks. Blystone in Texas Robert V. Blystone, Ph.D. Professor of Biology Trinity University San Antonio, Texas 78212 210.999-7243 210.999-7229 FAX -------------------------------- End of nih-image-d Digest V99 Issue #224 **************************************** From nih-image-request@io.ece.drexel.edu Fri Oct 1 14:55 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id OAA05161; Fri, 1 Oct 1999 14:55:02 -0400 (EDT) Resent-Date: Fri, 1 Oct 1999 14:55:02 -0400 (EDT) Message-ID: From: "Smiley, John" To: "'nih-image@biomed.drexel.edu'" Subject: archive Date: Fri, 1 Oct 1999 14:30:04 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2448.0) Resent-Message-ID: <"_yODZ.0.k37.usFzt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1781 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain Content-Length: 27 Status: O egrep live paste latest/* From nih-image-d-request@io.ece.drexel.edu Sat Oct 2 06:09 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA18807; Sat, 2 Oct 1999 06:09:31 -0400 (EDT) Date: Sat, 2 Oct 1999 06:09:31 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910021009.GAA18807@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #225 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/225 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 631 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 225 Today's Topics: archive [ "Smiley, John" To: "'nih-image@biomed.drexel.edu'" Subject: archive Message-ID: Content-Type: text/plain egrep live paste latest/* -------------------------------- End of nih-image-d Digest V99 Issue #225 **************************************** From nih-image-request@io.ece.drexel.edu Sat Oct 2 13:32 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA08529; Sat, 2 Oct 1999 13:32:11 -0400 (EDT) Resent-Date: Sat, 2 Oct 1999 13:32:11 -0400 (EDT) Message-Id: In-Reply-To: <199910021005.GAA17635@io.ece.drexel.edu> Mime-Version: 1.0 Date: Sat, 2 Oct 1999 19:23:12 +0200 To: nih-image@io.ece.drexel.edu From: Michael Glotzer Subject: import 16 bit images Resent-Message-ID: <"1locA2.0.SH1.NwZzt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1782 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 676 Status: O Imagers- I have a set of images from a time lapse series acquired with a 12 bit digital camera. I would like to import them all using the same scale to convert them to 8 bit. Using the import command (and "open all" checked), the images are each scaled differently regardless of how I set the calibrate and min max settings. I would be grateful for any suggestions. Michael ___________________________________________________________________ Michael Glotzer Research Institute of Molecular Pathology Dr. Bohr-Gasse 7 A-1030 Vienna AUSTRIA phone 43-(1)797-30-405 office phone 43-(1)797-30-525 lab fax 43-(1)798-7153 email mglotzer@nt.imp.univie.ac.at From nih-image-request@io.ece.drexel.edu Sat Oct 2 21:31 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id VAA26260; Sat, 2 Oct 1999 21:31:54 -0400 (EDT) Resent-Date: Sat, 2 Oct 1999 21:31:54 -0400 (EDT) Message-Id: <3.0.5.32.19991002212754.00a42500@codon.nih.gov> X-Sender: wayne@codon.nih.gov X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Sat, 02 Oct 1999 21:27:54 -0400 To: nih-image@io.ece.drexel.edu From: Wayne Rasband Subject: Re: import 16 bit images In-Reply-To: References: <199910021005.GAA17635@io.ece.drexel.edu> Mime-Version: 1.0 Resent-Message-ID: <"EL7mP1.0.Rf5.vygzt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1783 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 684 Status: O At 07:23 PM 10/2/99 +0200, you wrote: >Imagers- >I have a set of images from a time lapse series acquired with a 12 bit >digital camera. I would like to import them all using the same scale to >convert them to 8 bit. Using the import command (and "open all" checked), >the images are each scaled differently regardless of how I set the >calibrate and min max settings. I would be grateful for any suggestions. Use the File/Acquire/All as Stack command in ImageJ to open the images as a 16-bit stack. Convert the stack to 8-bits using the Image/Type/8-bit Grayscale command. Use File/Save As/Tiff to save the 8-bit stack in tiff format. Open the tiff stack in NIH Image. -wayne From nih-image-d-request@io.ece.drexel.edu Sun Oct 3 05:17 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA10802; Sun, 3 Oct 1999 05:17:06 -0400 (EDT) Date: Sun, 3 Oct 1999 05:17:06 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910030917.FAA10802@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #226 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/226 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 6377 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 226 Today's Topics: import 16 bit images [ Michael Glotzer ] ADMIN: list commands [ nih-image-owner@io.ece.drexel.edu ] ------------------------------ Date: Sat, 2 Oct 1999 19:23:12 +0200 From: Michael Glotzer To: nih-image@io.ece.drexel.edu Subject: import 16 bit images Message-Id: Content-Type: text/plain; charset="us-ascii" Imagers- I have a set of images from a time lapse series acquired with a 12 bit digital camera. I would like to import them all using the same scale to convert them to 8 bit. Using the import command (and "open all" checked), the images are each scaled differently regardless of how I set the calibrate and min max settings. I would be grateful for any suggestions. Michael ___________________________________________________________________ Michael Glotzer Research Institute of Molecular Pathology Dr. Bohr-Gasse 7 A-1030 Vienna AUSTRIA phone 43-(1)797-30-405 office phone 43-(1)797-30-525 lab fax 43-(1)798-7153 email mglotzer@nt.imp.univie.ac.at ------------------------------ Date: Sat, 02 Oct 1999 21:27:54 -0400 From: Wayne Rasband To: nih-image@io.ece.drexel.edu Subject: Re: import 16 bit images Message-Id: <3.0.5.32.19991002212754.00a42500@codon.nih.gov> Content-Type: text/plain; charset="us-ascii" At 07:23 PM 10/2/99 +0200, you wrote: >Imagers- >I have a set of images from a time lapse series acquired with a 12 bit >digital camera. I would like to import them all using the same scale to >convert them to 8 bit. Using the import command (and "open all" checked), >the images are each scaled differently regardless of how I set the >calibrate and min max settings. I would be grateful for any suggestions. Use the File/Acquire/All as Stack command in ImageJ to open the images as a 16-bit stack. Convert the stack to 8-bits using the Image/Type/8-bit Grayscale command. Use File/Save As/Tiff to save the 8-bit stack in tiff format. Open the tiff stack in NIH Image. -wayne ------------------------------ Date: Sun, 3 Oct 1999 05:05:01 -0400 (EDT) From: nih-image-owner@io.ece.drexel.edu To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Message-Id: <199910030905.FAA07872@io.ece.drexel.edu> NIH-image Mailing List Help --------------------------- nih-image-d-request@biomed.drexel.edu - Aministration for Digest nih-image-request@biomed.drexel.edu - Administation for List ********************************************************* * DO NOT SEND ADMINISTRATIVE COMMANDS TO THE LIST * ********************************************************* nih-image@biomed.drexel.edu - Sends mail to everyone on the list Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- -------------------------------- End of nih-image-d Digest V99 Issue #226 **************************************** From nih-image-request@io.ece.drexel.edu Sun Oct 3 05:21 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA12080; Sun, 3 Oct 1999 05:21:13 -0400 (EDT) Resent-Date: Sun, 3 Oct 1999 05:21:13 -0400 (EDT) Date: Sun, 3 Oct 1999 05:05:01 -0400 (EDT) From: nih-image-owner@io.ece.drexel.edu Message-Id: <199910030905.FAA07872@io.ece.drexel.edu> To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Resent-Message-ID: <"UbdQO3.0.Cx1._mnzt"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1784 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 3743 Status: O NIH-image Mailing List Help --------------------------- nih-image-d-request@biomed.drexel.edu - Aministration for Digest nih-image-request@biomed.drexel.edu - Administation for List ********************************************************* * DO NOT SEND ADMINISTRATIVE COMMANDS TO THE LIST * ********************************************************* nih-image@biomed.drexel.edu - Sends mail to everyone on the list Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- From nih-image-request@io.ece.drexel.edu Mon Oct 4 11:20 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA28191; Mon, 4 Oct 1999 11:20:10 -0400 (EDT) Resent-Date: Mon, 4 Oct 1999 11:20:10 -0400 (EDT) Message-Id: <3.0.5.32.19991004105728.00a4aae0@mailserver.aecom.yu.edu> X-Sender: cammer@mailserver.aecom.yu.edu X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Mon, 04 Oct 1999 10:57:28 -0700 To: nih-image@io.ece.drexel.edu From: Michael Cammer Subject: Re: import 16 bit images In-Reply-To: References: <199910021005.GAA17635@io.ece.drexel.edu> Mime-Version: 1.0 Resent-Message-ID: <"hadES2.0.Kz5.40C-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1785 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 970 Status: O We found this problem three years ago. What you need to do is write a script to openas for each one where the settings are reset for each one. Someone here is using such a script, but I don;t know where she has it saved. >I have a set of images from a time lapse series acquired with a 12 bit >digital camera. I would like to import them all using the same scale to >convert them to 8 bit. Using the import command (and "open all" checked), >the images are each scaled differently regardless of how I set the >calibrate and min max settings. > ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** From nih-image-request@io.ece.drexel.edu Mon Oct 4 11:22 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA28678; Mon, 4 Oct 1999 11:22:51 -0400 (EDT) Resent-Date: Mon, 4 Oct 1999 11:22:51 -0400 (EDT) Message-Id: <3.0.5.32.19991004110228.00a50a00@mailserver.aecom.yu.edu> X-Sender: cammer@mailserver.aecom.yu.edu X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Mon, 04 Oct 1999 11:02:28 -0700 To: nih-image@io.ece.drexel.edu From: Michael Cammer Subject: Re: Histogram. In-Reply-To: Mime-Version: 1.0 Resent-Message-ID: <"vIL-q1.0.vA6.h4C-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1786 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 963 Status: O Histogram values are stored in the histogram array. You should be able to set a range from a point on the downslopes of one peak to a point on the downslope on the next peak. Then something like this... for count := firstpoint to lastpoint do if count = firstpoint then minimum := histogram[count] else if histogram[count] < minimum then minimum := histogram[count]; >I have a histogram of an image showing three tops and I want to get the >pixelvalue of the minimal point between the second and third top. How >can I do that? ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 5 03:39 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id DAA28624; Tue, 5 Oct 1999 03:39:01 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 03:39:01 -0400 (EDT) Date: Tue, 05 Oct 1999 09:20:04 +0200 From: Lukas Schaerer Subject: CLSM In-reply-to: <199908110921.KAA16169@holyrood.ed.ac.uk> X-Sender: lschaere@ubecx01.unibe.ch To: nih-image@io.ece.drexel.edu Message-id: MIME-version: 1.0 Content-transfer-encoding: 7BIT Resent-Message-ID: <"Fuk721.0.T96.iQQ-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1787 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 1188 Status: O Hi Imagers, This probably is slightly off track, as it does not direcly relate to NIH-Image. But many of you use confocal laser scanning microscopy and maybe you can help me with this one. I would like to measure the volumes of internal structures of a copepod, namely the volume of a parasite larva that dwells in the body cavity, using stereology. The copepod carapax is about 1 x 0.5 x 0.5 mm and almost totally transparent. What I have previously done is histological sectioning, but this leads to problems of shrinkage (when using paraffin embedding) and is immensly time-consuming. An faster alternative, at least as it appears to me, could be to make optical sections with CLSM. I am, however, not sure how deep I will be able to penetrate the specimen. I guess it will depend on just how transparent it is, but I'm not sure. Any advice on that would be greatly appreciated, be it 'forget it', 'try it, it could work' or 'no problem, that is a sure bet'. Looking forward to hearing from you, Lukas Lukas Schaerer Abteilung Verhaltensoekologie Zoologisches Institut Universitaet Bern Wohlenstr. 50a 3032 Hinterkappelen Tel. +41 31 631 91 60 od. 16 Fax. +41 31 631 91 41 From nih-image-d-request@io.ece.drexel.edu Tue Oct 5 03:39 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id DAA28623; Tue, 5 Oct 1999 03:39:01 -0400 (EDT) Date: Tue, 5 Oct 1999 03:39:01 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910050739.DAA28623@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #227 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/227 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 4473 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 227 Today's Topics: Re: import 16 bit images [ Michael Cammer To: nih-image@io.ece.drexel.edu Subject: Re: import 16 bit images Message-Id: <3.0.5.32.19991004105728.00a4aae0@mailserver.aecom.yu.edu> Content-Type: text/plain; charset="us-ascii" We found this problem three years ago. What you need to do is write a script to openas for each one where the settings are reset for each one. Someone here is using such a script, but I don;t know where she has it saved. >I have a set of images from a time lapse series acquired with a 12 bit >digital camera. I would like to import them all using the same scale to >convert them to 8 bit. Using the import command (and "open all" checked), >the images are each scaled differently regardless of how I set the >calibrate and min max settings. > ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** ------------------------------ Date: Mon, 04 Oct 1999 11:02:28 -0700 From: Michael Cammer To: nih-image@io.ece.drexel.edu Subject: Re: Histogram. Message-Id: <3.0.5.32.19991004110228.00a50a00@mailserver.aecom.yu.edu> Content-Type: text/plain; charset="us-ascii" Histogram values are stored in the histogram array. You should be able to set a range from a point on the downslopes of one peak to a point on the downslope on the next peak. Then something like this... for count := firstpoint to lastpoint do if count = firstpoint then minimum := histogram[count] else if histogram[count] < minimum then minimum := histogram[count]; >I have a histogram of an image showing three tops and I want to get the >pixelvalue of the minimal point between the second and third top. How >can I do that? ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** ------------------------------ Date: Tue, 05 Oct 1999 09:20:04 +0200 From: Lukas Schaerer To: nih-image@io.ece.drexel.edu Subject: CLSM Message-id: Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT Hi Imagers, This probably is slightly off track, as it does not direcly relate to NIH-Image. But many of you use confocal laser scanning microscopy and maybe you can help me with this one. I would like to measure the volumes of internal structures of a copepod, namely the volume of a parasite larva that dwells in the body cavity, using stereology. The copepod carapax is about 1 x 0.5 x 0.5 mm and almost totally transparent. What I have previously done is histological sectioning, but this leads to problems of shrinkage (when using paraffin embedding) and is immensly time-consuming. An faster alternative, at least as it appears to me, could be to make optical sections with CLSM. I am, however, not sure how deep I will be able to penetrate the specimen. I guess it will depend on just how transparent it is, but I'm not sure. Any advice on that would be greatly appreciated, be it 'forget it', 'try it, it could work' or 'no problem, that is a sure bet'. Looking forward to hearing from you, Lukas Lukas Schaerer Abteilung Verhaltensoekologie Zoologisches Institut Universitaet Bern Wohlenstr. 50a 3032 Hinterkappelen Tel. +41 31 631 91 60 od. 16 Fax. +41 31 631 91 41 -------------------------------- End of nih-image-d Digest V99 Issue #227 **************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 5 04:51 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id EAA10136; Tue, 5 Oct 1999 04:51:18 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 04:51:18 -0400 (EDT) Date: Tue, 5 Oct 1999 01:30:13 -0700 (PDT) From: Stephan Anagnostaras To: nih-image@io.ece.drexel.edu Subject: Total # of observations In-Reply-To: <199910050734.DAA27592@io.ece.drexel.edu> Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"uRtoE2.0.5e1.8SR-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1788 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1537 Status: O Hello, I am using Image to do some video analysis. Basically the script takes 4 ROI's from a movie 300 frames long and gives me the S.D. for each frame. I have the max measurements set to 8000 and # of digits to the deciminal point set to 8. We have been using the script fine to get the 1200 observations per movie. The computer (using NIH image 1.62) has plenty of RAM (288 megs, Image gets about 210 megs, and the movies in question are something like 190 megs long). The problem i am having is now I want to increase the number of frames to 600. This would be 2400 observations which should be fine. But what happens is that when I hit 1280 observations Image freaks out and doesn't take any more observations (although the script continues to run normally). i have tried increasing the RAM (Image still has over 10 megs free even when the full 600-frame movie is loaded), and clipboard, etc. to no avail. Any ideas what I am missing? As I stated i have the dialog box for max observations set to 8000 correctly. The workaround right now is to do two separate analysis of 1200 observations each, but this is impractical for other reasons for us. Thanks, Stephan ------------------------------- Stephan G. Anagnostaras, Ph.D. UCLA Dept. of Neurobiology, Psychiatry, Psychology, Brain Research Institute, etc. stephan@lifesci.ucla.edu ------------------------------- Work, Gonda: (310) 267-2522 Psych: (310) 794-5339 Fax: (310) 206-5895 Home: (310) 306-0294 ------------------------------ From nih-image-request@io.ece.drexel.edu Tue Oct 5 05:32 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA16673; Tue, 5 Oct 1999 05:32:12 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 05:32:12 -0400 (EDT) Date: Tue, 5 Oct 1999 14:48:59 +0530 (IST) From: Ajai Vyas To: nih-image@io.ece.drexel.edu Subject: Filling in circular ROI. Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"pFMAZ.0.yM3.v5S-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1789 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1370 Status: O Hello, I am using ScionImage on 128 Mb computer for measuring total lengths of dendrites in neurons. I am interested in lengths in sections of defined radial distances from cell body. Macro then copies whole image it on a larger canvas, makes circular ROI's of increasing diameters and measure length successively. I need a larger canvas size because operations inside ROI demands that whole selection be inside image. But this makes the new image much larger to manage. The farthest distance I want to measure is some 800 pixels on either side of cell body. Hence image should be larger than 1600 pixels plus some additional 400, as cell body is not in centre of image many a times. One way is to scan images at lesser magnification. If somebody knows about any procedure to by-pass the need of ROI being entirely in image and yet do things like pixel count inside ROI, it will make the life easier. Thanks in advance, AJAI. ------------------------------------------------------------------------------- AJAI VYAS, National Center For Biological Sciences, Phone-080-8560459 UAS-GKVK Campus, 080-8561654 Bangalore-560065, INDIA. Ext. -3221 & 5009 E-mail - ajai@ncbs.res.in Fax - 080-8561662 - ajaivyas@hotmail.com ------------------------------------------------------------------------------- From nih-image-request@io.ece.drexel.edu Tue Oct 5 08:05 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA10551; Tue, 5 Oct 1999 08:05:45 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 08:05:45 -0400 (EDT) Date: Tue, 05 Oct 1999 13:44:34 +0200 From: Lukas Schaerer Subject: Re: CLSM In-reply-to: X-Sender: lschaere@ubecx01.unibe.ch To: nih-image@io.ece.drexel.edu Message-id: MIME-version: 1.0 Content-transfer-encoding: 7BIT References: <199908110921.KAA16169@holyrood.ed.ac.uk> Resent-Message-ID: <"JXAzN3.0.Kl1.MJU-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1790 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 630 Status: O Hi Imagers, I have a follow-up question to my previous post regarding confocal laser scanning microscopy (the additional questions always only pop up in my mind after pressing the send button). Is fluorescent staining unavoidable, or in other words, can unstained tissue generally be visualized in CLSM? May autofluorescence be sufficient to get a picture of the spatial arrangement of the structures (and if yes at which wavelength)? Thanks, Lukas Lukas Schaerer Abteilung Verhaltensoekologie Zoologisches Institut Universitaet Bern Wohlenstr. 50a 3032 Hinterkappelen Tel. +41 31 631 91 60 od. 16 Fax. +41 31 631 91 41 From nih-image-request@io.ece.drexel.edu Tue Oct 5 08:51 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA18376; Tue, 5 Oct 1999 08:51:43 -0400 (EDT) <199908110921.KAA16169@holyrood.ed.ac.uk> Resent-Date: Tue, 5 Oct 1999 08:51:43 -0400 (EDT) Message-Id: In-Reply-To: References: <199908110921.KAA16169@holyrood.ed.ac.uk> Mime-Version: 1.0 Date: Tue, 5 Oct 1999 07:37:44 -0500 To: nih-image@io.ece.drexel.edu From: Arnout Ruifrok Subject: Re: CLSM Resent-Message-ID: <"ZY2RY2.0.7m3.y3V-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1792 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 789 Status: O >Hi Imagers, > >I have a follow-up question to my previous post regarding confocal laser >scanning microscopy (the additional questions always only pop up in my mind >after pressing the send button). > >Is fluorescent staining unavoidable, or in other words, can unstained >tissue generally be visualized in CLSM? May autofluorescence be sufficient >to get a picture of the spatial arrangement of the structures (and if yes >at which wavelength)? > >Thanks, > > >Lukas > >Lukas Schaerer >Abteilung Verhaltensoekologie >Zoologisches Institut >Universitaet Bern >Wohlenstr. 50a >3032 Hinterkappelen > >Tel. +41 31 631 91 60 od. 16 >Fax. +41 31 631 91 41 Is your sample fixed? if you use glutaraldehide, your sample will probably light up like a christmas tree, in the green range Arnout From nih-image-request@io.ece.drexel.edu Tue Oct 5 08:52 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA18652; Tue, 5 Oct 1999 08:52:26 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 08:52:26 -0400 (EDT) Message-Id: In-Reply-To: References: <199908110921.KAA16169@holyrood.ed.ac.uk> Mime-Version: 1.0 Date: Tue, 5 Oct 1999 07:34:26 -0500 To: nih-image@io.ece.drexel.edu From: Arnout Ruifrok Subject: Re: CLSM Resent-Message-ID: <"T2EsA1.0.Pd3.s0V-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1791 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1429 Status: O Dear Lukas, What you metion should be doable. I have worked on a project were images were made of whole mounts of intestinal crypts, also in the mm range. So my answer is: try it, it should :) work. >Hi Imagers, > >This probably is slightly off track, as it does not direcly relate to >NIH-Image. But many of you use confocal laser scanning microscopy and maybe >you can help me with this one. > >I would like to measure the volumes of internal structures of a copepod, >namely the volume of a parasite larva that dwells in the body cavity, using >stereology. The copepod carapax is about 1 x 0.5 x 0.5 mm and almost >totally transparent. What I have previously done is histological >sectioning, but this leads to problems of shrinkage (when using paraffin >embedding) and is immensly time-consuming. > >An faster alternative, at least as it appears to me, could be to make >optical sections with CLSM. I am, however, not sure how deep I will be able >to penetrate the specimen. I guess it will depend on just how transparent >it is, but I'm not sure. > >Any advice on that would be greatly appreciated, be it 'forget it', 'try >it, it could work' or 'no problem, that is a sure bet'. > >Looking forward to hearing from you, > > >Lukas > > >Lukas Schaerer >Abteilung Verhaltensoekologie >Zoologisches Institut >Universitaet Bern >Wohlenstr. 50a >3032 Hinterkappelen > >Tel. +41 31 631 91 60 od. 16 >Fax. +41 31 631 91 41 From nih-image-request@io.ece.drexel.edu Tue Oct 5 09:10 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA22427; Tue, 5 Oct 1999 09:10:49 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 09:10:49 -0400 (EDT) Date: Tue, 05 Oct 1999 08:58:48 +0100 From: Julian Smith III Subject: Re: CLSM In-reply-to: To: nih-image@io.ece.drexel.edu Message-id: MIME-version: 1.0 References: <199908110921.KAA16169@holyrood.ed.ac.uk> Resent-Message-ID: <"5aiNG.0.1j4.ILV-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1793 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" <199908110921.KAA16169@holyrood.ed.ac.uk> Content-Length: 1044 Status: O Chelicerates are highly fluorescent, especially at UV wavelengths (must have been pretty sunny back when they crawled out on land....) I don't know about Crustaceans, but often just aldehyde fixation will induce weak autofluorescence. I'd vote for "try it and see". HTH Julian >I have a follow-up question to my previous post regarding confocal laser >scanning microscopy (the additional questions always only pop up in my mind >after pressing the send button). > >Is fluorescent staining unavoidable, or in other words, can unstained >tissue generally be visualized in CLSM? May autofluorescence be sufficient >to get a picture of the spatial arrangement of the structures (and if yes >at which wavelength)? > >Thanks, > > >Lukas > >Lukas Schaerer >Abteilung Verhaltensoekologie >Zoologisches Institut >Universitaet Bern >Wohlenstr. 50a >3032 Hinterkappelen > >Tel. +41 31 631 91 60 od. 16 >Fax. +41 31 631 91 41 Julian P.S. Smith III Dept. of Biology Winthrop University Rock Hill, SC 29733 803-323-2111 x6427 (vox) 803-323-3448 (fax) From nih-image-request@io.ece.drexel.edu Tue Oct 5 13:37 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA03570; Tue, 5 Oct 1999 13:37:51 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 13:37:51 -0400 (EDT) Message-Id: <3.0.5.32.19991005132253.0094ab80@mailserver.aecom.yu.edu> X-Sender: cammer@mailserver.aecom.yu.edu X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Tue, 05 Oct 1999 13:22:53 -0700 To: nih-image@io.ece.drexel.edu From: Michael Cammer Subject: Re: CLSM In-Reply-To: References: <199908110921.KAA16169@holyrood.ed.ac.uk> Mime-Version: 1.0 Resent-Message-ID: <"pW8do1.0.qC.DFZ-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1795 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 734 Status: O What about Nomarski optics to get a thin optical section? >>An faster alternative, at least as it appears to me, could be to make >>optical sections with CLSM. I am, however, not sure how deep I will be able >>to penetrate the specimen. I guess it will depend on just how transparent >>it is, but I'm not sure. ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 5 13:40 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA04097; Tue, 5 Oct 1999 13:40:22 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 13:40:22 -0400 (EDT) Message-Id: <3.0.5.32.19991005131841.00a495f0@mailserver.aecom.yu.edu> X-Sender: cammer@mailserver.aecom.yu.edu X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Tue, 05 Oct 1999 13:18:41 -0700 To: nih-image@io.ece.drexel.edu From: Michael Cammer Subject: Re: CLSM In-Reply-To: References: <199908110921.KAA16169@holyrood.ed.ac.uk> Mime-Version: 1.0 Resent-Message-ID: <"cmY2f.0.g3.EBZ-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1794 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 842 Status: O If you fix with glut, just about everything will light up, especially when excited at 488. We've also seen awesome excitation of connective tissue with multi-photon systems. >Is fluorescent staining unavoidable, or in other words, can unstained >tissue generally be visualized in CLSM? May autofluorescence be sufficient >to get a picture of the spatial arrangement of the structures (and if yes >at which wavelength)? ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 5 14:26 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id OAA11473; Tue, 5 Oct 1999 14:26:40 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 14:26:40 -0400 (EDT) Message-Id: In-Reply-To: References: <199910050734.DAA27592@io.ece.drexel.edu> Mime-Version: 1.0 Date: Tue, 5 Oct 1999 14:19:59 -0400 To: nih-image@io.ece.drexel.edu From: Wayne Rasband Subject: Re: Total # of observations Resent-Message-ID: <"Xwvog3.0.qD2.-xZ-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1796 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1358 Status: O >Hello, > >I am using Image to do some video analysis. Basically the script takes 4 >ROI's from a movie 300 frames long and gives me the S.D. for each frame. I >have the max measurements set to 8000 and # of digits to the deciminal >point set to 8. We have been using the script fine to get the 1200 >observations per movie. The computer (using NIH image 1.62) has plenty of >RAM (288 megs, Image gets about 210 megs, and the movies in question are >something like 190 megs long). > >The problem i am having is now I want to increase the number of frames to >600. This would be 2400 observations which should be fine. But what >happens is that when I hit 1280 observations Image freaks out and doesn't >take any more observations (although the script continues to run >normally). i have tried increasing the RAM (Image still has over 10 megs >free even when the full 600-frame movie is loaded), and clipboard, etc. to >no avail. Any ideas what I am missing? As I stated i have the dialog box >for max observations set to 8000 correctly. The workaround right now is >to do two separate analysis of 1200 observations each, but this is >impractical for other reasons for us. The Results window cannot display more than 32K of text but up to 8000 measuments are saved in RAM and can be exported to a tab-dilimeted text file using the Export command. -wayne From nih-image-request@io.ece.drexel.edu Tue Oct 5 16:17 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA26778; Tue, 5 Oct 1999 16:17:00 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 16:17:00 -0400 (EDT) Date: Tue, 5 Oct 1999 12:59:56 -0700 (PDT) From: "G. Macdonald" To: nih-image@io.ece.drexel.edu Subject: Matrox Millenium II for Mac In-Reply-To: <199910050741.DAA29020@io.ece.drexel.edu> Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"F0BC-.0.jx5.4Zb-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1797 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 510 Status: O has anyone successfully used a Matrox Millenium II PCI card on a Mac with a 50 MHz bus, such as a PowerMac 7300? I just obtained a surplused Millenium. Matrox thought it should work, even though it was designed for a 33 MHz bus. So far, I only get a black screen. Thanks, Glen Glen MacDonald Research Scientist Hearing Research Laboratories of the Virginia Merrill Bloedel Hearing Research Center Box 35-7923 University of Washington Seattle, WA 98195-7923 (206) 616-4156 glenmac@u.washington.edu From nih-image-request@io.ece.drexel.edu Tue Oct 5 18:10 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA13322; Tue, 5 Oct 1999 18:10:24 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 18:10:24 -0400 (EDT) Message-Id: In-Reply-To: References: <199910050741.DAA29020@io.ece.drexel.edu> Mime-Version: 1.0 Date: Tue, 5 Oct 1999 16:54:14 -0500 To: nih-image@io.ece.drexel.edu From: Arnout Ruifrok Subject: Re: Matrox Millenium II for Mac Resent-Message-ID: <"U-HBs.0.Sf2.jDd-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1798 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 589 Status: O >has anyone successfully used a Matrox Millenium II PCI card on a Mac with >a 50 MHz bus, such as a PowerMac 7300? I just obtained a surplused >Millenium. >Matrox thought it should work, even though it was designed for a 33 MHz >bus. So far, I only get a black screen. >Thanks, >Glen > > >Glen MacDonald > Research Scientist >Hearing Research Laboratories of the >Virginia Merrill Bloedel Hearing Research Center >Box 35-7923 >University of Washington >Seattle, WA 98195-7923 >(206) 616-4156 >glenmac@u.washington.edu Why the heck would you want to put a PC video card in a mac??? From nih-image-request@io.ece.drexel.edu Tue Oct 5 20:20 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id UAA03140; Tue, 5 Oct 1999 20:20:14 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 20:20:14 -0400 (EDT) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: ajai@ncbs.res.in, nih-image@io.ece.drexel.edu Date: Wed, 6 Oct 1999 10:13:30 +1000 Subject: Re:Filling in circular ROI(distance histogram Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: Resent-Message-ID: <"iiwp_.0.hB._Bf-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1799 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 3576 Status: O >Date: Tue, 5 Oct 1999 14:48:59 +0530 (IST) >From: Ajai Vyas >To: nih-image@io.ece.drexel.edu >Subject: Filling in circular ROI. > >Hello, >I am using ScionImage on 128 Mb computer for measuring total >lengths of dendrites in neurons. I am interested in lengths in sections >of defined radial distances from cell body. Macro then copies whole >image it on a larger canvas, makes circular ROI's of increasing >diameters and measure length successively. I need a larger canvas size >because operations inside ROI demands that whole selection be inside >image. But this makes the new image much larger to manage. The farthest >distance I want to measure is some 800 pixels on either side of cell >body. Hence image should be larger than 1600 pixels plus some >additional 400, as cell body is not in centre of image many a times. >One way is to scan images at lesser magnification. If somebody knows >about any procedure to by-pass the need of ROI being entirely in image >and yet do things like pixel count inside ROI, it will make the life >easier. >Thanks in advance, >AJAI. > >----------------------------------------------------------------------- > >AJAI VYAS, >National Center For Biological Sciences, Phone-080-8560459 >UAS-GKVK Campus, 080-8561654 >Bangalore-560065, INDIA. Ext. -3221 & 5009 >E-mail - ajai@ncbs.res.in Fax - 080-8561662 > - ajaivyas@hotmail.com > >----------------------------------------------------------------------- AJAI, No need to bother with circular ROI's. Make a template Euclidean Distance Map using procedure edmTemplate;begin setbackground(255); setNewSize(512,512); makeNewWindow('edmTemplate'); putPixel(255,255,0); binary('edm'); end macro'/0 edmTemplate';begin edmTemplate;end The result is an image with intensity graded 0>255 from centre radially outwards. The histogram is a wedge from 1 to 64K pixels giving the pixel count of the circumference of the circle at the corresponding intensity (radius). If you paste this image over the binary version of your dendrite image (or its skeleton) such that the centre of the image is at the centre of the cell body using the AND option in paste: macro'/1 measure'; var i,p,x,y:integer; begin p:=pidNumber; getMouse(x,y); {place cursor at cell body center and press <1>} selectWindow('edmTemplate'); selectAll; copy; selectPic(p); makeRoi(x-255,y-255,512,512); paste; setOption; doAND; selectAll;showHistogram; resetCounter; setCounter(255); setOptions('User1'); setUser1label('r:count'); for i:=1 to 255 do rUser1[i]:=histogram[i]; showResults; end then the histogram of the resultant image with give the pixel count at distances 0<>255 from the centre. The resolution is limited to 8 bit (256) but you can scale the template to match your original images. ie scaleandRotate(800/512,800/512,0); will give a template for a 800x800 image with 255 radius graduations. The paste operation does not require that the clipboard fall wholly within the image pasted to. Using a single value density slice and sliding the lut tool so that a red circle of the slider-controlled radius is displayed on the template will quickly verify its efficacy. The above works with NIH-Image. I assume that it would work with ScionImage except that for <1> use <1> for ScionImage. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-d-request@io.ece.drexel.edu Tue Oct 5 20:21 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id UAA03416; Tue, 5 Oct 1999 20:21:22 -0400 (EDT) Date: Tue, 5 Oct 1999 20:21:22 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910060021.UAA03416@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #228 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/228 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 18997 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 228 Today's Topics: Total # of observations [ Stephan Anagnostaras ] Re: CLSM [ Lukas Schaerer ] Matrox Millenium II for Mac [ "G. Macdonald" To: nih-image@io.ece.drexel.edu Subject: Total # of observations Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, I am using Image to do some video analysis. Basically the script takes 4 ROI's from a movie 300 frames long and gives me the S.D. for each frame. I have the max measurements set to 8000 and # of digits to the deciminal point set to 8. We have been using the script fine to get the 1200 observations per movie. The computer (using NIH image 1.62) has plenty of RAM (288 megs, Image gets about 210 megs, and the movies in question are something like 190 megs long). The problem i am having is now I want to increase the number of frames to 600. This would be 2400 observations which should be fine. But what happens is that when I hit 1280 observations Image freaks out and doesn't take any more observations (although the script continues to run normally). i have tried increasing the RAM (Image still has over 10 megs free even when the full 600-frame movie is loaded), and clipboard, etc. to no avail. Any ideas what I am missing? As I stated i have the dialog box for max observations set to 8000 correctly. The workaround right now is to do two separate analysis of 1200 observations each, but this is impractical for other reasons for us. Thanks, Stephan ------------------------------- Stephan G. Anagnostaras, Ph.D. UCLA Dept. of Neurobiology, Psychiatry, Psychology, Brain Research Institute, etc. stephan@lifesci.ucla.edu ------------------------------- Work, Gonda: (310) 267-2522 Psych: (310) 794-5339 Fax: (310) 206-5895 Home: (310) 306-0294 - ---------------------------- ------------------------------ Date: Tue, 5 Oct 1999 14:48:59 +0530 (IST) From: Ajai Vyas To: nih-image@io.ece.drexel.edu Subject: Filling in circular ROI. Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, I am using ScionImage on 128 Mb computer for measuring total lengths of dendrites in neurons. I am interested in lengths in sections of defined radial distances from cell body. Macro then copies whole image it on a larger canvas, makes circular ROI's of increasing diameters and measure length successively. I need a larger canvas size because operations inside ROI demands that whole selection be inside image. But this makes the new image much larger to manage. The farthest distance I want to measure is some 800 pixels on either side of cell body. Hence image should be larger than 1600 pixels plus some additional 400, as cell body is not in centre of image many a times. One way is to scan images at lesser magnification. If somebody knows about any procedure to by-pass the need of ROI being entirely in image and yet do things like pixel count inside ROI, it will make the life easier. Thanks in advance, AJAI. ------------------------------------------------------------------------------- AJAI VYAS, National Center For Biological Sciences, Phone-080-8560459 UAS-GKVK Campus, 080-8561654 Bangalore-560065, INDIA. Ext. -3221 & 5009 E-mail - ajai@ncbs.res.in Fax - 080-8561662 - ajaivyas@hotmail.com ------------------------------------------------------------------------------- ------------------------------ Date: Tue, 05 Oct 1999 13:44:34 +0200 From: Lukas Schaerer To: nih-image@io.ece.drexel.edu Subject: Re: CLSM Message-id: Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT Hi Imagers, I have a follow-up question to my previous post regarding confocal laser scanning microscopy (the additional questions always only pop up in my mind after pressing the send button). Is fluorescent staining unavoidable, or in other words, can unstained tissue generally be visualized in CLSM? May autofluorescence be sufficient to get a picture of the spatial arrangement of the structures (and if yes at which wavelength)? Thanks, Lukas Lukas Schaerer Abteilung Verhaltensoekologie Zoologisches Institut Universitaet Bern Wohlenstr. 50a 3032 Hinterkappelen Tel. +41 31 631 91 60 od. 16 Fax. +41 31 631 91 41 ------------------------------ Date: Tue, 5 Oct 1999 07:34:26 -0500 From: Arnout Ruifrok To: nih-image@io.ece.drexel.edu Subject: Re: CLSM Message-Id: Content-Type: text/plain; charset="us-ascii" Dear Lukas, What you metion should be doable. I have worked on a project were images were made of whole mounts of intestinal crypts, also in the mm range. So my answer is: try it, it should :) work. >Hi Imagers, > >This probably is slightly off track, as it does not direcly relate to >NIH-Image. But many of you use confocal laser scanning microscopy and maybe >you can help me with this one. > >I would like to measure the volumes of internal structures of a copepod, >namely the volume of a parasite larva that dwells in the body cavity, using >stereology. The copepod carapax is about 1 x 0.5 x 0.5 mm and almost >totally transparent. What I have previously done is histological >sectioning, but this leads to problems of shrinkage (when using paraffin >embedding) and is immensly time-consuming. > >An faster alternative, at least as it appears to me, could be to make >optical sections with CLSM. I am, however, not sure how deep I will be able >to penetrate the specimen. I guess it will depend on just how transparent >it is, but I'm not sure. > >Any advice on that would be greatly appreciated, be it 'forget it', 'try >it, it could work' or 'no problem, that is a sure bet'. > >Looking forward to hearing from you, > > >Lukas > > >Lukas Schaerer >Abteilung Verhaltensoekologie >Zoologisches Institut >Universitaet Bern >Wohlenstr. 50a >3032 Hinterkappelen > >Tel. +41 31 631 91 60 od. 16 >Fax. +41 31 631 91 41 ------------------------------ Date: Tue, 5 Oct 1999 07:37:44 -0500 From: Arnout Ruifrok To: nih-image@io.ece.drexel.edu Subject: Re: CLSM Message-Id: Content-Type: text/plain; charset="us-ascii" >Hi Imagers, > >I have a follow-up question to my previous post regarding confocal laser >scanning microscopy (the additional questions always only pop up in my mind >after pressing the send button). > >Is fluorescent staining unavoidable, or in other words, can unstained >tissue generally be visualized in CLSM? May autofluorescence be sufficient >to get a picture of the spatial arrangement of the structures (and if yes >at which wavelength)? > >Thanks, > > >Lukas > >Lukas Schaerer >Abteilung Verhaltensoekologie >Zoologisches Institut >Universitaet Bern >Wohlenstr. 50a >3032 Hinterkappelen > >Tel. +41 31 631 91 60 od. 16 >Fax. +41 31 631 91 41 Is your sample fixed? if you use glutaraldehide, your sample will probably light up like a christmas tree, in the green range Arnout ------------------------------ Date: Tue, 05 Oct 1999 08:58:48 +0100 From: Julian Smith III To: nih-image@io.ece.drexel.edu Subject: Re: CLSM Message-id: Content-type: text/plain; charset="us-ascii" Chelicerates are highly fluorescent, especially at UV wavelengths (must have been pretty sunny back when they crawled out on land....) I don't know about Crustaceans, but often just aldehyde fixation will induce weak autofluorescence. I'd vote for "try it and see". HTH Julian >I have a follow-up question to my previous post regarding confocal laser >scanning microscopy (the additional questions always only pop up in my mind >after pressing the send button). > >Is fluorescent staining unavoidable, or in other words, can unstained >tissue generally be visualized in CLSM? May autofluorescence be sufficient >to get a picture of the spatial arrangement of the structures (and if yes >at which wavelength)? > >Thanks, > > >Lukas > >Lukas Schaerer >Abteilung Verhaltensoekologie >Zoologisches Institut >Universitaet Bern >Wohlenstr. 50a >3032 Hinterkappelen > >Tel. +41 31 631 91 60 od. 16 >Fax. +41 31 631 91 41 Julian P.S. Smith III Dept. of Biology Winthrop University Rock Hill, SC 29733 803-323-2111 x6427 (vox) 803-323-3448 (fax) ------------------------------ Date: Tue, 05 Oct 1999 13:18:41 -0700 From: Michael Cammer To: nih-image@io.ece.drexel.edu Subject: Re: CLSM Message-Id: <3.0.5.32.19991005131841.00a495f0@mailserver.aecom.yu.edu> Content-Type: text/plain; charset="us-ascii" If you fix with glut, just about everything will light up, especially when excited at 488. We've also seen awesome excitation of connective tissue with multi-photon systems. >Is fluorescent staining unavoidable, or in other words, can unstained >tissue generally be visualized in CLSM? May autofluorescence be sufficient >to get a picture of the spatial arrangement of the structures (and if yes >at which wavelength)? ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** ------------------------------ Date: Tue, 05 Oct 1999 13:22:53 -0700 From: Michael Cammer To: nih-image@io.ece.drexel.edu Subject: Re: CLSM Message-Id: <3.0.5.32.19991005132253.0094ab80@mailserver.aecom.yu.edu> Content-Type: text/plain; charset="us-ascii" What about Nomarski optics to get a thin optical section? >>An faster alternative, at least as it appears to me, could be to make >>optical sections with CLSM. I am, however, not sure how deep I will be able >>to penetrate the specimen. I guess it will depend on just how transparent >>it is, but I'm not sure. ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** ------------------------------ Date: Tue, 5 Oct 1999 14:19:59 -0400 From: Wayne Rasband To: nih-image@io.ece.drexel.edu Subject: Re: Total # of observations Message-Id: Content-Type: text/plain; charset="us-ascii" >Hello, > >I am using Image to do some video analysis. Basically the script takes 4 >ROI's from a movie 300 frames long and gives me the S.D. for each frame. I >have the max measurements set to 8000 and # of digits to the deciminal >point set to 8. We have been using the script fine to get the 1200 >observations per movie. The computer (using NIH image 1.62) has plenty of >RAM (288 megs, Image gets about 210 megs, and the movies in question are >something like 190 megs long). > >The problem i am having is now I want to increase the number of frames to >600. This would be 2400 observations which should be fine. But what >happens is that when I hit 1280 observations Image freaks out and doesn't >take any more observations (although the script continues to run >normally). i have tried increasing the RAM (Image still has over 10 megs >free even when the full 600-frame movie is loaded), and clipboard, etc. to >no avail. Any ideas what I am missing? As I stated i have the dialog box >for max observations set to 8000 correctly. The workaround right now is >to do two separate analysis of 1200 observations each, but this is >impractical for other reasons for us. The Results window cannot display more than 32K of text but up to 8000 measuments are saved in RAM and can be exported to a tab-dilimeted text file using the Export command. -wayne ------------------------------ Date: Tue, 5 Oct 1999 12:59:56 -0700 (PDT) From: "G. Macdonald" To: nih-image@io.ece.drexel.edu Subject: Matrox Millenium II for Mac Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII has anyone successfully used a Matrox Millenium II PCI card on a Mac with a 50 MHz bus, such as a PowerMac 7300? I just obtained a surplused Millenium. Matrox thought it should work, even though it was designed for a 33 MHz bus. So far, I only get a black screen. Thanks, Glen Glen MacDonald Research Scientist Hearing Research Laboratories of the Virginia Merrill Bloedel Hearing Research Center Box 35-7923 University of Washington Seattle, WA 98195-7923 (206) 616-4156 glenmac@u.washington.edu ------------------------------ Date: Tue, 5 Oct 1999 16:54:14 -0500 From: Arnout Ruifrok To: nih-image@io.ece.drexel.edu Subject: Re: Matrox Millenium II for Mac Message-Id: Content-Type: text/plain; charset="us-ascii" >has anyone successfully used a Matrox Millenium II PCI card on a Mac with >a 50 MHz bus, such as a PowerMac 7300? I just obtained a surplused >Millenium. >Matrox thought it should work, even though it was designed for a 33 MHz >bus. So far, I only get a black screen. >Thanks, >Glen > > >Glen MacDonald > Research Scientist >Hearing Research Laboratories of the >Virginia Merrill Bloedel Hearing Research Center >Box 35-7923 >University of Washington >Seattle, WA 98195-7923 >(206) 616-4156 >glenmac@u.washington.edu Why the heck would you want to put a PC video card in a mac??? ------------------------------ Date: Wed, 6 Oct 1999 10:13:30 +1000 From: GJOSS@rna.bio.mq.edu.au To: ajai@ncbs.res.in, nih-image@io.ece.drexel.edu Subject: Re:Filling in circular ROI(distance histogram Message-ID: >Date: Tue, 5 Oct 1999 14:48:59 +0530 (IST) >From: Ajai Vyas >To: nih-image@io.ece.drexel.edu >Subject: Filling in circular ROI. > >Hello, >I am using ScionImage on 128 Mb computer for measuring total >lengths of dendrites in neurons. I am interested in lengths in sections >of defined radial distances from cell body. Macro then copies whole >image it on a larger canvas, makes circular ROI's of increasing >diameters and measure length successively. I need a larger canvas size >because operations inside ROI demands that whole selection be inside >image. But this makes the new image much larger to manage. The farthest >distance I want to measure is some 800 pixels on either side of cell >body. Hence image should be larger than 1600 pixels plus some >additional 400, as cell body is not in centre of image many a times. >One way is to scan images at lesser magnification. If somebody knows >about any procedure to by-pass the need of ROI being entirely in image >and yet do things like pixel count inside ROI, it will make the life >easier. >Thanks in advance, >AJAI. > >----------------------------------------------------------------------- > >AJAI VYAS, >National Center For Biological Sciences, Phone-080-8560459 >UAS-GKVK Campus, 080-8561654 >Bangalore-560065, INDIA. Ext. -3221 & 5009 >E-mail - ajai@ncbs.res.in Fax - 080-8561662 > - ajaivyas@hotmail.com > >----------------------------------------------------------------------- AJAI, No need to bother with circular ROI's. Make a template Euclidean Distance Map using procedure edmTemplate;begin setbackground(255); setNewSize(512,512); makeNewWindow('edmTemplate'); putPixel(255,255,0); binary('edm'); end macro'/0 edmTemplate';begin edmTemplate;end The result is an image with intensity graded 0>255 from centre radially outwards. The histogram is a wedge from 1 to 64K pixels giving the pixel count of the circumference of the circle at the corresponding intensity (radius). If you paste this image over the binary version of your dendrite image (or its skeleton) such that the centre of the image is at the centre of the cell body using the AND option in paste: macro'/1 measure'; var i,p,x,y:integer; begin p:=pidNumber; getMouse(x,y); {place cursor at cell body center and press <1>} selectWindow('edmTemplate'); selectAll; copy; selectPic(p); makeRoi(x-255,y-255,512,512); paste; setOption; doAND; selectAll;showHistogram; resetCounter; setCounter(255); setOptions('User1'); setUser1label('r:count'); for i:=1 to 255 do rUser1[i]:=histogram[i]; showResults; end then the histogram of the resultant image with give the pixel count at distances 0<>255 from the centre. The resolution is limited to 8 bit (256) but you can scale the template to match your original images. ie scaleandRotate(800/512,800/512,0); will give a template for a 800x800 image with 255 radius graduations. The paste operation does not require that the clipboard fall wholly within the image pasted to. Using a single value density slice and sliding the lut tool so that a red circle of the slider-controlled radius is displayed on the template will quickly verify its efficacy. The above works with NIH-Image. I assume that it would work with ScionImage except that for <1> use <1> for ScionImage. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia -------------------------------- End of nih-image-d Digest V99 Issue #228 **************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 5 21:53 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id VAA18313; Tue, 5 Oct 1999 21:53:31 -0400 (EDT) Resent-Date: Tue, 5 Oct 1999 21:53:31 -0400 (EDT) Date: Tue, 5 Oct 1999 18:41:56 -0700 (PDT) From: "G. Macdonald" To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #228 In-Reply-To: <199910060021.UAA03467@io.ece.drexel.edu> Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"DqnvZ2.0.H_3.gZg-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1800 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 650 Status: O >Why the heck would you want to put a PC video card in a mac??? Are you referring to the company or the bus? Matrox made a version of their Millenium card for the Mac a couple of years ago. Didn't hear much about it at the time. The PCI bus is the current standard for expansion cards in both Mac and Wintel computers. I'm actually not so much interested in the video out as in the video in capabilities. Glen Glen MacDonald Research Scientist Hearing Research Laboratories of the Virginia Merrill Bloedel Hearing Research Center Box 35-7923 University of Washington Seattle, WA 98195-7923 (206) 616-4156 glenmac@u.washington.edu From nih-image-request@io.ece.drexel.edu Wed Oct 6 01:57 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id BAA26877; Wed, 6 Oct 1999 01:57:52 -0400 (EDT) Resent-Date: Wed, 6 Oct 1999 01:57:52 -0400 (EDT) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: gary.chinga@pfi.no, nih-image@io.ece.drexel.edu, Date: Wed, 6 Oct 1999 15:52:20 +1000 Subject: Histogram analysis for thresholding/segmentation Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: Resent-Message-ID: <"wu18L.0.166.P9k-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1801 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 2738 Status: O I was interested in Gary Chinga's post re Histogram in that I was hoping to learn something useful from some of the more knowledgeable list participants re formal and well structured analysis of histograms for the purpose of choosing thresholds for segmentation of images. Michael Cammer's response would be useful at the level of referencing macro code/facilities which allow access to histogram values but relies on operator provision of the range over which to search for a minima. It concentrated on simplicity and ,probably quite reasonably, did not address pattern recognition aspects or how to locate cardinal points ie maxima, minima or zero crossings of derivatives and ignored problems associated with noise. Apart from the algorithm built into NIH-Image as "AutoThreshold;", I dont know of any useful alternative or more general algorithms although I see occasional references to unspecified texts on image analysis for more formal approaches to histogram analysis. In cases where I have had occasion to do more than autothreshold, I have generally resorted to a variety of hacks on the histogram array, smoothing, differenceing, find zero crossings etc and then progressive test of cardinal points against some heuristic (ie estimated guess/fudge factors) to automate a predetermined segmentation based on experiments with variations on expected images. If someone has code or a reference to more formal and better structured analysis to address this process, I would be grateful for a post to this list (or direct to myself if prefered). I recall making a post to the old maillist some years ago giving my interpretation of the mechanism NIH-Image "AutoThreshold" algorithm but I have to admit that I haven't looked into ImageJ to see if Wayne has used the same contributed algorithm as he did in NIH-Image. To address Gary's problem of segmenting the image between the second and third peaks of the histogram, the following code is effective with images whose histogram conforms to the general characteristics of Gary's case. It works without the need for operator bracketing of the search range by finding and eliminating the image component responsible for the first peak (ie left-hand-side or low valued peak) in the histogram and then having a second byte at the cherry. macro'/0 2rd minima';var lower,upper:integer; begin duplicate('temp'); selectAll; autoThreshold; getThresholds(lower,upper); changeValues(0,lower,0); autoThreshold; getThresholds(lower,upper); dispose; setThreshold(lower); end Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Wed Oct 6 03:45 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id DAA14051; Wed, 6 Oct 1999 03:45:05 -0400 (EDT) <199908110921.KAA16169@holyrood.ed.ac.uk> Resent-Date: Wed, 6 Oct 1999 03:45:05 -0400 (EDT) X-Sender: fzhao@pop.helsinki.fi Message-Id: In-Reply-To: References: <199908110921.KAA16169@holyrood.ed.ac.uk> Mime-Version: 1.0 Date: Wed, 6 Oct 1999 10:34:24 +0300 To: nih-image@io.ece.drexel.edu From: Fang Zhao Subject: Re: CLSM Resent-Message-ID: <"0kWzw3.0.dr2.Jhl-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1802 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 440 Status: O >Is fluorescent staining unavoidable, or in other words, can unstained >tissue generally be visualized in CLSM? May autofluorescence be sufficient >to get a picture of the spatial arrangement of the structures (and if yes >at which wavelength)? hi, Lukas: Fluorescent is not unavoidable. What brand of CLSM are you using? You should be able to choose transmission mode and use phase contrast or DIC if you have such auxiliaries. Fang From nih-image-request@io.ece.drexel.edu Wed Oct 6 11:19 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA23452; Wed, 6 Oct 1999 11:19:37 -0400 (EDT) Resent-Date: Wed, 6 Oct 1999 11:19:37 -0400 (EDT) Subject: frame capture Date: Wed, 6 Oct 99 11:00:10 -0400 x-sender: jlr$biol@qc1.qc.edu x-mailer: Claris Emailer 1.1 From: jared rifkin To: "nih-image users group" Mime-Version: 1.0 Message-ID: <6EE7415102@qc1.qc.edu> Resent-Message-ID: <"R_DAn3.0.Kz4.aGs-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1803 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 1078 Status: O i am a novice to nih-image so please reply with patience. and- many thanks in advance for any help. i need to trap single frames of black&white video-microscopic images with very high resolution (at least 1000x1000 pixel) to generate time-lapse series for time/motion-vector study. the images appear great on my tv monitor and i can feed either analog (ntsc) or digital video to my pc (i have access to both a firewire-mac and a windows pc with hi-res capture board. question 1: does nih-image work with digital video input? (firewire connection) question 2: if an analog signal is required- does nih-image work in windows environment? question 3: if either q1 or q2 has a "yes" answer- then (and-remember i'm a total novice to nih-image) how hard is it to write a script that will trap one frame every n seconds for m hours? manual start is fine- even manual stop is ok. i would welcome extended discussions so please email me directly. i'm in new york city and would welcome phone help. dr. jared l. rifkin biology department queens college 718-997-3432 From nih-image-request@io.ece.drexel.edu Wed Oct 6 11:54 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA28976; Wed, 6 Oct 1999 11:54:58 -0400 (EDT) Resent-Date: Wed, 6 Oct 1999 11:54:58 -0400 (EDT) Message-Id: X-Mailer: Novell GroupWise 5.2 Date: Wed, 06 Oct 1999 11:37:08 -0400 From: "Wade Schuette" To: cammer@aecom.yu.edu, nih-image@io.ece.drexel.edu, gary.chinga@pfi.no, GJOSS@rna.bio.mq.edu.au Subject: Re: Histogram analysis for thresholding/segmentation Mime-Version: 1.0 Content-Disposition: inline Resent-Message-ID: <"hyN_s1.0.jO6.tos-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1804 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=US-ASCII Content-Length: 2910 Status: O Hi, Greg! Two thoughts on this issue. The first, that I'm sure I've already posted on in the archives somewhere, is the idea that, should any scientific measurement be contingent on the threshold, then one better assess the dependence on threshold and determine if the hypothesis (conclusion) is true over the entire range of potential thresholds. If not, one should treat the conclusion as unsupported. (This of course requires measuring images multiple times, and requires automation to be sensible, but it avoids making a whole class of terrible mistakes and wasting time chasing false leads.) The second is that, for thin section biological images, the histogram contains information that is dependent upon both the biology and the slice thickness, so it is NOT a conclusion-neutral activity to set the histogram. For example, if cells with LARGE nuclei and small nucelei are mixed on a nominal "4-micron" slice, the LARGE ones will be selectively sliced proportionately thinner, so any stain that is constant per nuclei will tend to miss large nuclei, which is quite bad if one is trying to determine, say ratio of large to small nuclei. You may be interested in my thought paper, "Model-Based Estimation of Specimen Thickness on pathology Slides from the stain intensity histogram", online as the first paper (WS99.001) on the list at: http://www-personal.umich.edu/~schuette/imaging (one needs the Acrobat PDF reader to view the document.) In that paper I simulate various thickness virtual slices from distributions of spheres and look at what the histograms look like. There are some distinctive cliffs and features of the histograms that are reflections of the slice thickness, which may vary from slide to slide, and certainly from day to day or operator to operator. Under some conditions sometimes obtained in reality, the thickness of the slice can actually be determined from the intensity histogram shape and features. (I did this because we discovered that "uniform" thickness slices varied from 4micron to about 20 microns, as well as that the actual thickness cut by the microtome is about twice the value shown on the dial.) There are very strong statistical implications of the thickness on measurements of similarly sized objects, such as nuclei. Wade Schuette University of Michigan Medical Center Ann Arbor, MI USA >>> 10/06 1:57 AM >>> I was interested in Gary Chinga's post re Histogram in that I was hoping to learn something useful from some of the more knowledgeable list participants re formal and well structured analysis of histograms for the purpose of choosing thresholds for segmentation of images. ...If someone has code or a reference to more formal and better structured analysis to address this process, I would be grateful for a post to this list (or direct to myself if prefered). ... Greg Joss, From nih-image-request@io.ece.drexel.edu Wed Oct 6 16:01 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA03993; Wed, 6 Oct 1999 16:01:46 -0400 (EDT) Resent-Date: Wed, 6 Oct 1999 16:01:46 -0400 (EDT) Date: Wed, 06 Oct 1999 12:42:13 -0700 From: David Linker To: jared rifkin cc: nih-image@io.ece.drexel.edu Subject: Re: frame capture Message-ID: <139118.3148202533@huginn.medicine.washington.edu> In-Reply-To: <6EE7415102@qc1.qc.edu> Originator-Info: login-id=dtlinker; server=dtlinker.deskmail.washington.edu X-Mailer: Mulberry (MacOS) [1.4.0, s/n U-300938] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline Resent-Message-ID: <"ttasG1.0.CA.UOw-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1805 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 1758 Status: O 1. NIH-Image has support for digitizing from video using any hardware that has a vdig component. That includes essentially all Mac-specific digitization hardware. I do not know how this works with firewire. 2. NIH-Image does not work under Windows. 3. It would be quite easy (probably <10 min) to write a macro to capture a sequence of images, if you have enough memory to store all of the images prior to saving them. If you want "incremental saving", it is a little more difficult, and depends on how you want to save the images. DTL --On Wed, Oct 6, 1999 11:00 AM -0400 jared rifkin wrote: > i am a novice to nih-image so please reply with patience. and- many > thanks in advance for any help. > > i need to trap single frames of black&white video-microscopic images with > very high resolution (at least 1000x1000 pixel) to generate time-lapse > series for time/motion-vector study. the images appear great on my tv > monitor and i can feed either analog (ntsc) or digital video to my pc (i > have access to both a firewire-mac and a windows pc with hi-res capture > board. > > question 1: does nih-image work with digital video input? (firewire > connection) > question 2: if an analog signal is required- does nih-image work in > windows environment? > question 3: if either q1 or q2 has a "yes" answer- then (and-remember i'm > a total novice to nih-image) how hard is it to write a script that will > trap one frame every n seconds for m hours? manual start is fine- even > manual stop is ok. > > i would welcome extended discussions so please email me directly. i'm in > new york city and would welcome phone help. > > dr. jared l. rifkin > biology department > queens college > 718-997-3432 > > From nih-image-d-request@io.ece.drexel.edu Wed Oct 6 16:02 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA04089; Wed, 6 Oct 1999 16:02:12 -0400 (EDT) Date: Wed, 6 Oct 1999 16:02:12 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910062002.QAA04089@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #229 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/229 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 12178 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 229 Today's Topics: Re: nih-image-d Digest V99 #228 [ "G. Macdonald" ] frame capture [ jared rifkin ] Re: Histogram analysis for threshold [ "Wade Schuette" To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #228 Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII >Why the heck would you want to put a PC video card in a mac??? Are you referring to the company or the bus? Matrox made a version of their Millenium card for the Mac a couple of years ago. Didn't hear much about it at the time. The PCI bus is the current standard for expansion cards in both Mac and Wintel computers. I'm actually not so much interested in the video out as in the video in capabilities. Glen Glen MacDonald Research Scientist Hearing Research Laboratories of the Virginia Merrill Bloedel Hearing Research Center Box 35-7923 University of Washington Seattle, WA 98195-7923 (206) 616-4156 glenmac@u.washington.edu ------------------------------ Date: Wed, 6 Oct 1999 15:52:20 +1000 From: GJOSS@rna.bio.mq.edu.au To: gary.chinga@pfi.no, nih-image@io.ece.drexel.edu, Subject: Histogram analysis for thresholding/segmentation Message-ID: I was interested in Gary Chinga's post re Histogram in that I was hoping to learn something useful from some of the more knowledgeable list participants re formal and well structured analysis of histograms for the purpose of choosing thresholds for segmentation of images. Michael Cammer's response would be useful at the level of referencing macro code/facilities which allow access to histogram values but relies on operator provision of the range over which to search for a minima. It concentrated on simplicity and ,probably quite reasonably, did not address pattern recognition aspects or how to locate cardinal points ie maxima, minima or zero crossings of derivatives and ignored problems associated with noise. Apart from the algorithm built into NIH-Image as "AutoThreshold;", I dont know of any useful alternative or more general algorithms although I see occasional references to unspecified texts on image analysis for more formal approaches to histogram analysis. In cases where I have had occasion to do more than autothreshold, I have generally resorted to a variety of hacks on the histogram array, smoothing, differenceing, find zero crossings etc and then progressive test of cardinal points against some heuristic (ie estimated guess/fudge factors) to automate a predetermined segmentation based on experiments with variations on expected images. If someone has code or a reference to more formal and better structured analysis to address this process, I would be grateful for a post to this list (or direct to myself if prefered). I recall making a post to the old maillist some years ago giving my interpretation of the mechanism NIH-Image "AutoThreshold" algorithm but I have to admit that I haven't looked into ImageJ to see if Wayne has used the same contributed algorithm as he did in NIH-Image. To address Gary's problem of segmenting the image between the second and third peaks of the histogram, the following code is effective with images whose histogram conforms to the general characteristics of Gary's case. It works without the need for operator bracketing of the search range by finding and eliminating the image component responsible for the first peak (ie left-hand-side or low valued peak) in the histogram and then having a second byte at the cherry. macro'/0 2rd minima';var lower,upper:integer; begin duplicate('temp'); selectAll; autoThreshold; getThresholds(lower,upper); changeValues(0,lower,0); autoThreshold; getThresholds(lower,upper); dispose; setThreshold(lower); end Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Wed, 6 Oct 1999 10:34:24 +0300 From: Fang Zhao To: nih-image@io.ece.drexel.edu Subject: Re: CLSM Message-Id: Content-Type: text/plain; charset="us-ascii" >Is fluorescent staining unavoidable, or in other words, can unstained >tissue generally be visualized in CLSM? May autofluorescence be sufficient >to get a picture of the spatial arrangement of the structures (and if yes >at which wavelength)? hi, Lukas: Fluorescent is not unavoidable. What brand of CLSM are you using? You should be able to choose transmission mode and use phase contrast or DIC if you have such auxiliaries. Fang ------------------------------ Date: Wed, 6 Oct 99 11:00:10 -0400 From: jared rifkin To: "nih-image users group" Subject: frame capture Message-ID: <6EE7415102@qc1.qc.edu> Content-Type: text/plain; charset="US-ASCII" i am a novice to nih-image so please reply with patience. and- many thanks in advance for any help. i need to trap single frames of black&white video-microscopic images with very high resolution (at least 1000x1000 pixel) to generate time-lapse series for time/motion-vector study. the images appear great on my tv monitor and i can feed either analog (ntsc) or digital video to my pc (i have access to both a firewire-mac and a windows pc with hi-res capture board. question 1: does nih-image work with digital video input? (firewire connection) question 2: if an analog signal is required- does nih-image work in windows environment? question 3: if either q1 or q2 has a "yes" answer- then (and-remember i'm a total novice to nih-image) how hard is it to write a script that will trap one frame every n seconds for m hours? manual start is fine- even manual stop is ok. i would welcome extended discussions so please email me directly. i'm in new york city and would welcome phone help. dr. jared l. rifkin biology department queens college 718-997-3432 ------------------------------ Date: Wed, 06 Oct 1999 11:37:08 -0400 From: "Wade Schuette" To: cammer@aecom.yu.edu, nih-image@io.ece.drexel.edu, gary.chinga@pfi.no, GJOSS@rna.bio.mq.edu.au Subject: Re: Histogram analysis for thresholding/segmentation Message-Id: Content-Type: text/plain; charset=US-ASCII Content-Disposition: inline Hi, Greg! Two thoughts on this issue. The first, that I'm sure I've already posted on in the archives somewhere, is the idea that, should any scientific measurement be contingent on the threshold, then one better assess the dependence on threshold and determine if the hypothesis (conclusion) is true over the entire range of potential thresholds. If not, one should treat the conclusion as unsupported. (This of course requires measuring images multiple times, and requires automation to be sensible, but it avoids making a whole class of terrible mistakes and wasting time chasing false leads.) The second is that, for thin section biological images, the histogram contains information that is dependent upon both the biology and the slice thickness, so it is NOT a conclusion-neutral activity to set the histogram. For example, if cells with LARGE nuclei and small nucelei are mixed on a nominal "4-micron" slice, the LARGE ones will be selectively sliced proportionately thinner, so any stain that is constant per nuclei will tend to miss large nuclei, which is quite bad if one is trying to determine, say ratio of large to small nuclei. You may be interested in my thought paper, "Model-Based Estimation of Specimen Thickness on pathology Slides from the stain intensity histogram", online as the first paper (WS99.001) on the list at: http://www-personal.umich.edu/~schuette/imaging (one needs the Acrobat PDF reader to view the document.) In that paper I simulate various thickness virtual slices from distributions of spheres and look at what the histograms look like. There are some distinctive cliffs and features of the histograms that are reflections of the slice thickness, which may vary from slide to slide, and certainly from day to day or operator to operator. Under some conditions sometimes obtained in reality, the thickness of the slice can actually be determined from the intensity histogram shape and features. (I did this because we discovered that "uniform" thickness slices varied from 4micron to about 20 microns, as well as that the actual thickness cut by the microtome is about twice the value shown on the dial.) There are very strong statistical implications of the thickness on measurements of similarly sized objects, such as nuclei. Wade Schuette University of Michigan Medical Center Ann Arbor, MI USA >>> 10/06 1:57 AM >>> I was interested in Gary Chinga's post re Histogram in that I was hoping to learn something useful from some of the more knowledgeable list participants re formal and well structured analysis of histograms for the purpose of choosing thresholds for segmentation of images. ...If someone has code or a reference to more formal and better structured analysis to address this process, I would be grateful for a post to this list (or direct to myself if prefered). ... Greg Joss, ------------------------------ Date: Wed, 06 Oct 1999 12:42:13 -0700 From: David Linker To: jared rifkin cc: nih-image@io.ece.drexel.edu Subject: Re: frame capture Message-ID: <139118.3148202533@huginn.medicine.washington.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline 1. NIH-Image has support for digitizing from video using any hardware that has a vdig component. That includes essentially all Mac-specific digitization hardware. I do not know how this works with firewire. 2. NIH-Image does not work under Windows. 3. It would be quite easy (probably <10 min) to write a macro to capture a sequence of images, if you have enough memory to store all of the images prior to saving them. If you want "incremental saving", it is a little more difficult, and depends on how you want to save the images. DTL --On Wed, Oct 6, 1999 11:00 AM -0400 jared rifkin wrote: > i am a novice to nih-image so please reply with patience. and- many > thanks in advance for any help. > > i need to trap single frames of black&white video-microscopic images with > very high resolution (at least 1000x1000 pixel) to generate time-lapse > series for time/motion-vector study. the images appear great on my tv > monitor and i can feed either analog (ntsc) or digital video to my pc (i > have access to both a firewire-mac and a windows pc with hi-res capture > board. > > question 1: does nih-image work with digital video input? (firewire > connection) > question 2: if an analog signal is required- does nih-image work in > windows environment? > question 3: if either q1 or q2 has a "yes" answer- then (and-remember i'm > a total novice to nih-image) how hard is it to write a script that will > trap one frame every n seconds for m hours? manual start is fine- even > manual stop is ok. > > i would welcome extended discussions so please email me directly. i'm in > new york city and would welcome phone help. > > dr. jared l. rifkin > biology department > queens college > 718-997-3432 > > -------------------------------- End of nih-image-d Digest V99 Issue #229 **************************************** From nih-image-request@io.ece.drexel.edu Wed Oct 6 16:40 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA11827; Wed, 6 Oct 1999 16:40:11 -0400 (EDT) Resent-Date: Wed, 6 Oct 1999 16:40:11 -0400 (EDT) Message-Id: <3.0.5.32.19991006162635.0092c8f0@mailserver.aecom.yu.edu> X-Sender: cammer@mailserver.aecom.yu.edu X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Wed, 06 Oct 1999 16:26:35 -0700 To: nih-image@io.ece.drexel.edu From: Michael Cammer Subject: RE: Mac vs. PC In-Reply-To: References: < <28C8B514497ED111AF510000F803473E0105CA29@power.coe.montana.edu> Mime-Version: 1.0 Resent-Message-ID: <"f1mi11.0.W52.G1x-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1806 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 583 Status: O The fact is that Java is not stable or cross-platform compatible. Rgeardless of the hype. NIH-Image is a truly awesome program, but Image J is not there yet. ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** From nih-image-request@io.ece.drexel.edu Wed Oct 6 19:18 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA13301; Wed, 6 Oct 1999 19:18:05 -0400 (EDT) Resent-Date: Wed, 6 Oct 1999 19:18:05 -0400 (EDT) X-Sender: kdc17@mailhubl.cc.columbia.edu Message-Id: Mime-Version: 1.0 Date: Wed, 6 Oct 1999 19:03:48 -0700 To: nih-image@io.ece.drexel.edu From: "Kevin D. Costa" Subject: Re: frame capture Resent-Message-ID: <"OlYGk1.0.Mc2.sLz-t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1807 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1889 Status: O Here's a macro I downloaded from somewhere which allows intermittent frame grabbing and saves to disk. Depending on your video card, you may need to replace "Acquire('Plug-in Digitizer');" with "Capture;". I have used this macro with NIH Image on a mac...my understanding is that scion image (ftp://codon.nih.gov/pub/nih-image/nih-image_spin-offs/) provides similar functionality on the pc. ============================================================================== Kevin D. Costa, PhD Assistant Professor Department of Biomedical Engineering 212 854-9163 (phone) Columbia University 212 854-8725 (fax) 416 CEPSR, mail code 8904 530 W. 120th Street kdc17@columbia.edu New York, NY 10027 ============================================================================== macro 'Make AV Movie to Disk'; { Captures images using 'Plug-in Digitizer' and saves them to disk. Abort at any time by pressing the mouse button. } var nFrames,n:integer; interval,StartTicks,EndTicks:integer; time:real; path:string; begin Requiresversion(1.55); path := GetString('Folder path:','MyHD:movie'); nFrames := GetNumber('Number of Frames?',10); time := GetNumber('Delay Between Frames (seconds)?',60.0); interval := round(time*60); StartTicks := TickCount; EndTicks := TickCount+interval; for n := 1 to nFrames do begin time := (TickCount-StartTicks)/60; ShowMessage(n:3,' ',time:4:2); Acquire('Plug-in Digitizer'); MoveTo(2,12); SetFontSize(12); SetForegroundColor(255); write(n:3,' ',time:4:2); SaveAs(path,':Frame ',n); Dispose; while TickCount < EndTicks do begin if button then exit; end; EndTicks := EndTicks+interval; end; end; From nih-image-request@io.ece.drexel.edu Thu Oct 7 04:48 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id EAA12121; Thu, 7 Oct 1999 04:48:32 -0400 (EDT) <28C8B514497ED111AF510000F803473E0105CA29@power.coe.montana.edu> <3.0.5.32.19991006162635.0092c8f0@mailserver.aecom.yu.edu> Resent-Date: Thu, 7 Oct 1999 04:48:32 -0400 (EDT) Mime-Version: 1.0 X-Sender: lis3@pop.lis.ch Message-Id: In-Reply-To: <3.0.5.32.19991006162635.0092c8f0@mailserver.aecom.yu.edu> References: < <28C8B514497ED111AF510000F803473E0105CA29@power.coe.montana.edu> <3.0.5.32.19991006162635.0092c8f0@mailserver.aecom.yu.edu> Date: Thu, 7 Oct 1999 10:11:53 +0200 To: nih-image@io.ece.drexel.edu From: Beat Ludin Subject: RE: Mac vs. PC Resent-Message-ID: <"91TIA1.0.ux1.1X5_t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1808 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" ; format="flowed" Content-Length: 1532 Status: RO >The fact is that Java is not stable or cross-platform compatible. >Rgeardless of the hype. What's your definition of stability and compatibility? I'm using a pretty big piece of telebanking software written entirely in Java. Runs flawlessly under MacOS, Win95/98/NT and, from what I hear, various UNIX flavors. Tested ImageJ under MacOS and Win98 and it ran just fine. Both applications have been perfectly stable so far. I would say Java is as cross- platform compatible and stable as the virtual machine running it and the underlying OS - and the VMs have become pretty good lately (make sure you run the latest version available). Java is a more programmer-friendly language than, for instance, C++, too. As long as the guys in Redmond are not succeeding in screwing up the standard with their "embrace&extend" strategy, Java is probably here to stay and grow. Regardless of the neg-hype :-) > NIH-Image is a truly awesome program, but Image J is not there yet. Well, ImageJ does a few things already that NIH-Image does not, and I'm sure Wayne is working day and night to implement the missing features. Just my two dark current electrons Beat LIFE IMAGING SERVICES - visit our web site at http://www.lis.ch +----------------------------------------------------------- | Dr. B. Ludin | Life Imaging Services fon ++41 (0)79 235 7154 | Muehletalweg 22 fax ++41 (0)86 062 296 3160 NEW! | CH-4600 Olten beat.ludin@lis.ch | Switzerland http://www.lis.ch From nih-image-request@io.ece.drexel.edu Thu Oct 7 11:58 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA22040; Thu, 7 Oct 1999 11:58:30 -0400 (EDT) Resent-Date: Thu, 7 Oct 1999 11:58:30 -0400 (EDT) X-Sender: wgr2@pop.cus.cam.ac.uk Message-Id: Mime-Version: 1.0 Date: Thu, 7 Oct 1999 16:39:26 +0100 To: nih-image@io.ece.drexel.edu From: Dr Gareth Rees Subject: Publishing/publicising macros Resent-Message-ID: <"6gSWW.0.VB4.avB_t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1809 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1324 Status: O Hello imagers, Advice wanted! Partly for my own work (mostly satellite image processing), and sometimes just for fun, I write quite a few Image macros to do all sorts of image processing operations (random examples: terrain visualisation; geometric rectification; majority-filtering; automatic thresholding...). I pass these on to colleagues if they are interested in them (I quite often develop them for colleagues in the first place), and I am gradually starting to put them on a web site, but does anyone out there have other suggestions on how to make them more widely known? I did try submitting a couple of articles to journals, but the referees basically took the view that if a particular functionality is widely available already, even if only in expensive image processing suites, then there's no point in publishing implementations in Image. I disagree with this attitude for two reasons: (1) not everyone can afford the latest version of Erdas Imagine, or whatever; (2) if I can see how a macro has been put together, there's some chance that I can actually understand what it does! I am not saying that I have some radically wonderful new image processing algorithms (I don't - but I find them useful myself), but if anyone is sympathetic to my reasons (1) and (2) and has any bright ideas... Gareth Rees From nih-image-d-request@io.ece.drexel.edu Thu Oct 7 11:58 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA22059; Thu, 7 Oct 1999 11:58:36 -0400 (EDT) Date: Thu, 7 Oct 1999 11:58:36 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910071558.LAA22059@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #230 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/230 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 6982 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 230 Today's Topics: RE: Mac vs. PC [ Michael Cammer ] Publishing/publicising macros [ Dr Gareth Rees ] ------------------------------ Date: Wed, 06 Oct 1999 16:26:35 -0700 From: Michael Cammer To: nih-image@io.ece.drexel.edu Subject: RE: Mac vs. PC Message-Id: <3.0.5.32.19991006162635.0092c8f0@mailserver.aecom.yu.edu> Content-Type: text/plain; charset="us-ascii" The fact is that Java is not stable or cross-platform compatible. Rgeardless of the hype. NIH-Image is a truly awesome program, but Image J is not there yet. ******************************************************************** * Michael Cammer * Analytical Imaging Facility * Albert Einstein * * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * * personal URL: http://cammer.home.mindspring.com/ * ******************************************************************** ------------------------------ Date: Wed, 6 Oct 1999 19:03:48 -0700 From: "Kevin D. Costa" To: nih-image@io.ece.drexel.edu Subject: Re: frame capture Message-Id: Content-Type: text/plain; charset="us-ascii" Here's a macro I downloaded from somewhere which allows intermittent frame grabbing and saves to disk. Depending on your video card, you may need to replace "Acquire('Plug-in Digitizer');" with "Capture;". I have used this macro with NIH Image on a mac...my understanding is that scion image (ftp://codon.nih.gov/pub/nih-image/nih-image_spin-offs/) provides similar functionality on the pc. ============================================================================== Kevin D. Costa, PhD Assistant Professor Department of Biomedical Engineering 212 854-9163 (phone) Columbia University 212 854-8725 (fax) 416 CEPSR, mail code 8904 530 W. 120th Street kdc17@columbia.edu New York, NY 10027 ============================================================================== macro 'Make AV Movie to Disk'; { Captures images using 'Plug-in Digitizer' and saves them to disk. Abort at any time by pressing the mouse button. } var nFrames,n:integer; interval,StartTicks,EndTicks:integer; time:real; path:string; begin Requiresversion(1.55); path := GetString('Folder path:','MyHD:movie'); nFrames := GetNumber('Number of Frames?',10); time := GetNumber('Delay Between Frames (seconds)?',60.0); interval := round(time*60); StartTicks := TickCount; EndTicks := TickCount+interval; for n := 1 to nFrames do begin time := (TickCount-StartTicks)/60; ShowMessage(n:3,' ',time:4:2); Acquire('Plug-in Digitizer'); MoveTo(2,12); SetFontSize(12); SetForegroundColor(255); write(n:3,' ',time:4:2); SaveAs(path,':Frame ',n); Dispose; while TickCount < EndTicks do begin if button then exit; end; EndTicks := EndTicks+interval; end; end; ------------------------------ Date: Thu, 7 Oct 1999 10:11:53 +0200 From: Beat Ludin To: nih-image@io.ece.drexel.edu Subject: RE: Mac vs. PC Message-Id: Content-Type: text/plain; charset="us-ascii" ; format="flowed" >The fact is that Java is not stable or cross-platform compatible. >Rgeardless of the hype. What's your definition of stability and compatibility? I'm using a pretty big piece of telebanking software written entirely in Java. Runs flawlessly under MacOS, Win95/98/NT and, from what I hear, various UNIX flavors. Tested ImageJ under MacOS and Win98 and it ran just fine. Both applications have been perfectly stable so far. I would say Java is as cross- platform compatible and stable as the virtual machine running it and the underlying OS - and the VMs have become pretty good lately (make sure you run the latest version available). Java is a more programmer-friendly language than, for instance, C++, too. As long as the guys in Redmond are not succeeding in screwing up the standard with their "embrace&extend" strategy, Java is probably here to stay and grow. Regardless of the neg-hype :-) > NIH-Image is a truly awesome program, but Image J is not there yet. Well, ImageJ does a few things already that NIH-Image does not, and I'm sure Wayne is working day and night to implement the missing features. Just my two dark current electrons Beat LIFE IMAGING SERVICES - visit our web site at http://www.lis.ch +----------------------------------------------------------- | Dr. B. Ludin | Life Imaging Services fon ++41 (0)79 235 7154 | Muehletalweg 22 fax ++41 (0)86 062 296 3160 NEW! | CH-4600 Olten beat.ludin@lis.ch | Switzerland http://www.lis.ch ------------------------------ Date: Thu, 7 Oct 1999 16:39:26 +0100 From: Dr Gareth Rees To: nih-image@io.ece.drexel.edu Subject: Publishing/publicising macros Message-Id: Content-Type: text/plain; charset="us-ascii" Hello imagers, Advice wanted! Partly for my own work (mostly satellite image processing), and sometimes just for fun, I write quite a few Image macros to do all sorts of image processing operations (random examples: terrain visualisation; geometric rectification; majority-filtering; automatic thresholding...). I pass these on to colleagues if they are interested in them (I quite often develop them for colleagues in the first place), and I am gradually starting to put them on a web site, but does anyone out there have other suggestions on how to make them more widely known? I did try submitting a couple of articles to journals, but the referees basically took the view that if a particular functionality is widely available already, even if only in expensive image processing suites, then there's no point in publishing implementations in Image. I disagree with this attitude for two reasons: (1) not everyone can afford the latest version of Erdas Imagine, or whatever; (2) if I can see how a macro has been put together, there's some chance that I can actually understand what it does! I am not saying that I have some radically wonderful new image processing algorithms (I don't - but I find them useful myself), but if anyone is sympathetic to my reasons (1) and (2) and has any bright ideas... Gareth Rees -------------------------------- End of nih-image-d Digest V99 Issue #230 **************************************** From nih-image-request@io.ece.drexel.edu Thu Oct 7 13:14 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA06121; Thu, 7 Oct 1999 13:14:20 -0400 (EDT) Resent-Date: Thu, 7 Oct 1999 13:14:20 -0400 (EDT) Date: Thu, 07 Oct 1999 12:53:39 -0400 From: "Barry R. Bickmore" Subject: Re: Publishing/publicising macros In-reply-to: X-Sender: bbickmor@mail.vt.edu To: nih-image@io.ece.drexel.edu Message-id: MIME-version: 1.0 Resent-Message-ID: <"ZT41e.0._b.T0D_t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1810 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 2424 Status: O >Hello imagers, > >Advice wanted! > >Partly for my own work (mostly satellite image processing), and sometimes >just for fun, I write quite a few Image macros to do all sorts of image >processing operations (random examples: terrain visualisation; geometric >rectification; majority-filtering; automatic thresholding...). I pass these >on to colleagues if they are interested in them (I quite often develop them >for colleagues in the first place), and I am gradually starting to put them >on a web site, but does anyone out there have other suggestions on how to >make them more widely known? I did try submitting a couple of articles to >journals, but the referees basically took the view that if a particular >functionality is widely available already, even if only in expensive image >processing suites, then there's no point in publishing implementations in >Image. I disagree with this attitude for two reasons: (1) not everyone can >afford the latest version of Erdas Imagine, or whatever; (2) if I can see >how a macro has been put together, there's some chance that I can actually >understand what it does! > >I am not saying that I have some radically wonderful new image processing >algorithms (I don't - but I find them useful myself), but if anyone is >sympathetic to my reasons (1) and (2) and has any bright ideas... > >Gareth Rees They are doing a special issue of the Journal of Computer-Assisted Microscopy on scanning force microscopy image analysis. Several of us who write SFM macros for Image SXM (a spin off of NIH Image for SEM, SAM, STM, and SFM) are getting together to write an article on customized image analysis with this program. I have a web site to house the codes. Anyway, maybe a general article like that would pass review, rather than a specific article on one particular macro application. Barry _____________________________________________________________________________ Barry Bickmore "Lisa, just because Dept. of Geological Sciences I don't care doesn't mean Virginia Tech I don't understand!" Blacksburg, VA 24061 -Homer J. Simpson bbickmor@vt.edu (540)231-8575 FAX:(540)231-3386 Barry's preprints/reprints page: http://www.geocities.com/Athens/Parthenon/2671/preprint.html _____________________________________________________________________________ess From nih-image-request@io.ece.drexel.edu Thu Oct 7 15:30 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id PAA25764; Thu, 7 Oct 1999 15:30:25 -0400 (EDT) Resent-Date: Thu, 7 Oct 1999 15:30:25 -0400 (EDT) Message-ID: <37FCF039.43AE3962@ucsd.edu> Date: Thu, 07 Oct 1999 12:10:49 -0700 From: "Harvey J. Karten" Reply-To: hjkarten@ucsd.edu Organization: UCSD X-Mailer: Mozilla 4.04 [en]C-DIAL (WinNT; U) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Mac vs. PC NIH-Image and IMageJ References: <199910071556.LAA21647@io.ece.drexel.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"6rkrt1.0.Lf5.92F_t"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1811 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 3816 Status: O As a long time user of NIH-Image, I now find that ImageJ has some really powerful funtions that can't be replicated on NIH-Image. These include the extensive capabilities for dealing with 16 bit multichannel, multi-LUT (8 bit imposed LUT on the 16 bit images), that are really superb. There are constraints in the Java INterface that are less smooth thatn NIH-Image. I hope that those things get ironed out. In our working with ImageJ, we find that it is very robust on both Mac and PC. Even if occasional problems come up, it never crashes the whole damned system the way the bloody Microsoft programs do. Are you sure that you are running the correct versions of Java for each platform, QuickTime, etc. Hats off to Wayne. He again is going against the mainstream, and proving himself to be correct. Get on board and help us develop the software for the 21st Century. regards, Harvey Karten UCSD > Subject: RE: Mac vs. PC > Date: Wed, 06 Oct 1999 16:26:35 -0700 > From: Michael Cammer > To: nih-image@io.ece.drexel.edu > > The fact is that Java is not stable or cross-platform compatible. > Rgeardless of the hype. > NIH-Image is a truly awesome program, but Image J is not there yet. > > ******************************************************************** > * Michael Cammer * Analytical Imaging Facility * Albert Einstein * > * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * > * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * > * personal URL: http://cammer.home.mindspring.com/ * > ******************************************************************** > ------------------------------------------------------------------------ > > Subject: RE: Mac vs. PC > Date: Thu, 7 Oct 1999 10:11:53 +0200 > From: Beat Ludin > To: nih-image@io.ece.drexel.edu > > >The fact is that Java is not stable or cross-platform compatible. > >Rgeardless of the hype. > > What's your definition of stability and compatibility? I'm using a > pretty big piece of telebanking software written entirely in Java. > Runs flawlessly under MacOS, Win95/98/NT and, from what I hear, > various UNIX flavors. Tested ImageJ under MacOS and Win98 and it ran > just fine. Both applications have been perfectly stable so far. I > would say Java is as cross- platform compatible and stable as the > virtual machine running it and the underlying OS - and the VMs have > become pretty good lately (make sure you run the latest version > available). Java is a more programmer-friendly language than, for > instance, C++, too. As long as the guys in Redmond are not succeeding > in screwing up the standard with their "embrace&extend" strategy, > Java is probably here to stay and grow. Regardless of the neg-hype :-) > > > NIH-Image is a truly awesome program, but Image J is not there yet. > > Well, ImageJ does a few things already that NIH-Image does not, and > I'm sure Wayne is working day and night to implement the missing > features. > > Just my two dark current electrons > > Beat > > LIFE IMAGING SERVICES > - visit our web site at http://www.lis.ch > +----------------------------------------------------------- > | Dr. B. Ludin > | Life Imaging Services fon ++41 (0)79 235 7154 > | Muehletalweg 22 fax ++41 (0)86 062 296 3160 NEW! > | CH-4600 Olten beat.ludin@lis.ch > | Switzerland http://www.lis.ch > > ------------------------------------------------------------------------ > -- Harvey J. Karten, M.D. Dept. of Neurosciences University of California at San Diego La Jolla, CA 92093 Phone (Voice) 619-534-4938 FAX 619-534-6602 EMail: hjkarten@ucsd.edu Retina Information Home Page URL: http://www-cajal.ucsd.edu Tayana Manual: ftp://ftp.sailnet.com/lists/tayana/t-37manual/ From nih-image-request@io.ece.drexel.edu Thu Oct 7 16:27 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA03596; Thu, 7 Oct 1999 16:27:23 -0400 (EDT) Resent-Date: Thu, 7 Oct 1999 16:27:23 -0400 (EDT) Message-Id: <37FCFE4E.DAF0F222@maroon.tc.umn.edu> Date: Thu, 07 Oct 1999 15:10:55 -0500 From: "Michael J. Herron" Reply-To: Michael J Herron Organization: University of Minnesota X-Mailer: Mozilla 4.5 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: hjkarten@ucsd.edu CC: nih-image@io.ece.drexel.edu Subject: Re: Mac vs. PC NIH-Image and IMageJ References: <199910071556.LAA21647@io.ece.drexel.edu> <37FCF039.43AE3962@ucsd.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"oOeMW1.0.XE.5uF_t"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1812 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 4648 Status: O I was complaining to Wayne a couple of weeks ago that ImageJ was crashing my machine. I have since found that it works fine if I run it as a standalone applet instead of trying to run it over the web. I concur ImageJ has some great features esp. for confocalers. Is there any way to use ImageJ to grab from a trusty old Scion LG3? "Harvey J. Karten" wrote: > > As a long time user of NIH-Image, I now find that ImageJ has some really powerful > funtions that can't be replicated on NIH-Image. These include the extensive > capabilities for dealing with 16 bit multichannel, multi-LUT (8 bit imposed LUT on > the 16 bit images), that are really superb. There are constraints in the Java > INterface that are less smooth thatn NIH-Image. I hope that those things get ironed > out. In our working with ImageJ, we find that it is very robust on both Mac and PC. > Even if occasional problems come up, it never crashes the whole damned system the > way the bloody Microsoft programs do. Are you sure that you are running the correct > versions of Java for each platform, QuickTime, etc. > > Hats off to Wayne. He again is going against the mainstream, and proving himself to > be correct. Get on board and help us develop the software for the 21st Century. > > regards, Harvey Karten > UCSD > > > Subject: RE: Mac vs. PC > > Date: Wed, 06 Oct 1999 16:26:35 -0700 > > From: Michael Cammer > > To: nih-image@io.ece.drexel.edu > > > > The fact is that Java is not stable or cross-platform compatible. > > Rgeardless of the hype. > > NIH-Image is a truly awesome program, but Image J is not there yet. > > > > ******************************************************************** > > * Michael Cammer * Analytical Imaging Facility * Albert Einstein * > > * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * > > * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * > > * personal URL: http://cammer.home.mindspring.com/ * > > ******************************************************************** > > ------------------------------------------------------------------------ > > > > Subject: RE: Mac vs. PC > > Date: Thu, 7 Oct 1999 10:11:53 +0200 > > From: Beat Ludin > > To: nih-image@io.ece.drexel.edu > > > > >The fact is that Java is not stable or cross-platform compatible. > > >Rgeardless of the hype. > > > > What's your definition of stability and compatibility? I'm using a > > pretty big piece of telebanking software written entirely in Java. > > Runs flawlessly under MacOS, Win95/98/NT and, from what I hear, > > various UNIX flavors. Tested ImageJ under MacOS and Win98 and it ran > > just fine. Both applications have been perfectly stable so far. I > > would say Java is as cross- platform compatible and stable as the > > virtual machine running it and the underlying OS - and the VMs have > > become pretty good lately (make sure you run the latest version > > available). Java is a more programmer-friendly language than, for > > instance, C++, too. As long as the guys in Redmond are not succeeding > > in screwing up the standard with their "embrace&extend" strategy, > > Java is probably here to stay and grow. Regardless of the neg-hype :-) > > > > > NIH-Image is a truly awesome program, but Image J is not there yet. > > > > Well, ImageJ does a few things already that NIH-Image does not, and > > I'm sure Wayne is working day and night to implement the missing > > features. > > > > Just my two dark current electrons > > > > Beat > > > > LIFE IMAGING SERVICES > > - visit our web site at http://www.lis.ch > > +----------------------------------------------------------- > > | Dr. B. Ludin > > | Life Imaging Services fon ++41 (0)79 235 7154 > > | Muehletalweg 22 fax ++41 (0)86 062 296 3160 NEW! > > | CH-4600 Olten beat.ludin@lis.ch > > | Switzerland http://www.lis.ch > > > > ------------------------------------------------------------------------ > > > > -- > Harvey J. Karten, M.D. > Dept. of Neurosciences > University of California at San Diego > La Jolla, CA 92093 > Phone (Voice) 619-534-4938 > FAX 619-534-6602 > EMail: hjkarten@ucsd.edu > Retina Information Home Page > URL: http://www-cajal.ucsd.edu > Tayana Manual: ftp://ftp.sailnet.com/lists/tayana/t-37manual/ -- _______________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /__________________________________________/ From nih-image-request@io.ece.drexel.edu Thu Oct 7 17:02 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id RAA08627; Thu, 7 Oct 1999 17:02:12 -0400 (EDT) <37FCF039.43AE3962@ucsd.edu> Resent-Date: Thu, 7 Oct 1999 17:02:12 -0400 (EDT) Message-Id: In-Reply-To: <37FCFE4E.DAF0F222@maroon.tc.umn.edu> References: <199910071556.LAA21647@io.ece.drexel.edu> <37FCF039.43AE3962@ucsd.edu> Mime-Version: 1.0 Date: Thu, 7 Oct 1999 16:57:26 -0400 To: nih-image@io.ece.drexel.edu From: Wayne Rasband Subject: Re: Mac vs. PC NIH-Image and IMageJ Resent-Message-ID: <"ceYlo1.0.PY1.cRG_t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1813 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 860 Status: O >Is there any way to use ImageJ to grab from a trusty old Scion LG3? This is on the ImageJ to-do list, which I have attached. -wayne 1. Particle analysis. 2. Interactive thresholding and density slicing (bilevel thresholds). 3. Keyboard shortcuts for plug-ins. 4. A "mark and count" tool similar to the one in NIH Image. 5. Scripting language similar to NIH Image's Pascal-like macro language. 6. Display measurement results in a spreadsheet-like table with no arbitrary size limits. 7. Plug-ins that use QuickTime for Java to read and write QuickTime movies. 8. Surface plot plug-in. 9. Density calibration using the Simplex fitting method (straight line, polynomial, exponential, power and log fits) used by NIH image. 10. Fourier transforms. 11. On-line help. 12. Scion LG-3 frame grabber acquisition plug-in. 13. User modifiable convolution kernels. From nih-image-request@io.ece.drexel.edu Thu Oct 7 19:38 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA08058; Thu, 7 Oct 1999 19:38:24 -0400 (EDT) Resent-Date: Thu, 7 Oct 1999 19:38:24 -0400 (EDT) X-Sender: gjones@poserver-g.nih.gov Message-Id: In-Reply-To: <199910071559.LAA22315@io.ece.drexel.edu> Mime-Version: 1.0 Date: Thu, 7 Oct 1999 19:25:19 -0400 To: nih-image@io.ece.drexel.edu From: g jones Subject: Re: Publishing/publicising macros Resent-Message-ID: <"Lq-yF.0.nK1.YlI_t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1814 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 2039 Status: O >Date: Thu, 7 Oct 1999 16:39:26 +0100 >From: Dr Gareth Rees >To: nih-image@io.ece.drexel.edu >Subject: Publishing/publicising macros >Message-Id: >Content-Type: text/plain; charset="us-ascii" >MIME-Version: 1.0 > >Hello imagers, > >Advice wanted! > >Partly for my own work (mostly satellite image processing), and sometimes >just for fun, I write quite a few Image macros to do all sorts of image >processing operations (random examples: terrain visualisation; geometric >rectification; majority-filtering; automatic thresholding...). I pass these >on to colleagues if they are interested in them (I quite often develop them >for colleagues in the first place), and I am gradually starting to put them >on a web site, but does anyone out there have other suggestions on how to >make them more widely known? I did try submitting a couple of articles to >journals, but the referees basically took the view that if a particular >functionality is widely available already, even if only in expensive image >processing suites, then there's no point in publishing implementations in >Image. I disagree with this attitude for two reasons: (1) not everyone can >afford the latest version of Erdas Imagine, or whatever; (2) if I can see >how a macro has been put together, there's some chance that I can actually >understand what it does! > >I am not saying that I have some radically wonderful new image processing >algorithms (I don't - but I find them useful myself), but if anyone is >sympathetic to my reasons (1) and (2) and has any bright ideas... > >Gareth Rees > If nothing else, send them to this forum with a sparing description since they'll be archived and searchable... someday even by mere mortals with a friendlier interface instead of just the unix-speak we have now. Those of us who use image are automatically those who don't spend big bucks on software. Others don't need to care and won't see them anyway. Maybe Wayne will even like them and keep them on zippy. From nih-image-request@io.ece.drexel.edu Thu Oct 7 21:53 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id VAA02035; Thu, 7 Oct 1999 21:53:59 -0400 (EDT) Resent-Date: Thu, 7 Oct 1999 21:53:59 -0400 (EDT) User-Agent: Microsoft Outlook Express Macintosh Edition - 5.0 (1429.1) Date: Fri, 08 Oct 1999 11:41:41 +1000 Subject: Re: Publishing/publicising macros From: Timothy Bates To: Dr Gareth Rees , Message-ID: In-Reply-To: Mime-version: 1.0 Content-transfer-encoding: 7bit Resent-Message-ID: <"YadzT2.0.SI7.plK_t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1815 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 1814 Status: O Putting them on a web site and then mentioning that from time to time, where appropriate, on lists such as this, newsgroups etc will soon get you widely known. Another good idea is to put a button on the web page to allow people to subscribe to an updates mailing list. A good example is Denis Pelli's video toolbox site. There are now hundreds of users of his procedures and the site is very well known. Once established, a resource such as this might well warrant an article in BRM&C. Another example is the Tarr and Williams RSVP page. These things can grow to foster a community of users. Thus I wrote a support page for RSVP: And now Lawrence D'Oliviero and I are are developing an AppleScriptable derivative of RSVP called PsyScript. Just get the web page up and your audience is approximately 1000,000 times larger than ANY journal can hope to offer you. Plus no one is going to read the paper journal beyond 6months whereas the web page is finger-tip close. best, tim Thereon 10/8/99 1:39 AM, Dr Gareth Rees wrote: > Hello imagers, > Partly for my own work (mostly satellite image processing), and sometimes > just for fun, I write quite a few Image macros to do all sorts of image > processing operations (random examples: terrain visualisation; geometric > rectification; majority-filtering; automatic thresholding...). I pass these > on to colleagues if they are interested in them (I quite often develop them > for colleagues in the first place), and I am gradually starting to put them > on a web site, but does anyone out there have other suggestions on how to > make them more widely known? From nih-image-request@io.ece.drexel.edu Thu Oct 7 23:58 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id XAA25192; Thu, 7 Oct 1999 23:58:02 -0400 (EDT) Resent-Date: Thu, 7 Oct 1999 23:58:02 -0400 (EDT) Message-ID: <37FD69D8.C738E002@thurston.com> Date: Thu, 07 Oct 1999 20:49:50 -0700 From: Pat and Leslie Pringle Reply-To: lespat@thurston.com X-Mailer: Mozilla 4.05 (Macintosh; I; PPC) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #230 References: <199910071540.LAA17744@io.ece.drexel.edu> Content-Transfer-Encoding: 8bit Resent-Message-ID: <"CEsqb.0.1Y5.6aM_t"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1816 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=iso-8859-1; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Length: 2269 Status: O Hello imagers, re: Gareth Rees's comments, which included the following: "...I am gradually starting to put them (macros) on a web site, but does anyone out there have other suggestions on how to make them more widely known? I did try submitting a couple of articles to journals, but the referees basically took the view that if a particular functionality is widely available already, even if only in expensive image processing suites, then there's no point in publishing implementations in Image..." Well, I would suggest that perhaps certain publishers often have a different agenda than you or I—and they can change that agenda, or those tendencies, from year to year, depending on editors, megatrends, sunspots, and El Nino. Perhaps you should wait it out and submit again later. Then again, what would we researchers do without teamwork and cooperation such as you are describing in your email as your motivation for producing macros--so aiming at different (less widely known) publication media probably is advisable. Should we presuppose that all good science is focused on particular blue chipjournals?—NOT Plenty of people are doing good science with the most minimal of tools. It's people such as you, Wayne, Norbert Vischer, Greg Joss, Wayne Schuette, Ard, Beat, and numerous others who are making this possible. The collective "we" is stupendously grateful for your contributions. Consider that there are plenty of "top" scientists who neglect (or refuse?) even to put their supporting raw data in repositories or on the web even after they've published in blue chip journals. In this case, the :"great science" is more often than not self supporting with a healthy dose of "trust me", and not necessarily constructive to the whole edifice of scientific progress. You do, of course have, to worry about your own security and vitality as a scientist, so publication ("publish or perish") may be a must and may the Force be with you. But my suggestion would be to not underestimate the power of lower wattage journals or mediums (web sites) for much of what you do, while at the same time continuing your search for the right location having greater visibility and prestige. Let us know what the web page is when you're ready for that. Cheers, Pat Pringle From nih-image-d-request@io.ece.drexel.edu Fri Oct 8 06:11 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA06046; Fri, 8 Oct 1999 06:11:36 -0400 (EDT) Date: Fri, 8 Oct 1999 06:11:36 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910081011.GAA06046@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #231 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/231 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 20881 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 231 Today's Topics: Re: Publishing/publicising macros [ "Barry R. Bickmore" ] Re: Publishing/publicising macros [ g jones ] Re: Publishing/publicising macros [ Timothy Bates To: nih-image@io.ece.drexel.edu Subject: Re: Publishing/publicising macros Message-id: Content-type: text/plain; charset=us-ascii >Hello imagers, > >Advice wanted! > >Partly for my own work (mostly satellite image processing), and sometimes >just for fun, I write quite a few Image macros to do all sorts of image >processing operations (random examples: terrain visualisation; geometric >rectification; majority-filtering; automatic thresholding...). I pass these >on to colleagues if they are interested in them (I quite often develop them >for colleagues in the first place), and I am gradually starting to put them >on a web site, but does anyone out there have other suggestions on how to >make them more widely known? I did try submitting a couple of articles to >journals, but the referees basically took the view that if a particular >functionality is widely available already, even if only in expensive image >processing suites, then there's no point in publishing implementations in >Image. I disagree with this attitude for two reasons: (1) not everyone can >afford the latest version of Erdas Imagine, or whatever; (2) if I can see >how a macro has been put together, there's some chance that I can actually >understand what it does! > >I am not saying that I have some radically wonderful new image processing >algorithms (I don't - but I find them useful myself), but if anyone is >sympathetic to my reasons (1) and (2) and has any bright ideas... > >Gareth Rees They are doing a special issue of the Journal of Computer-Assisted Microscopy on scanning force microscopy image analysis. Several of us who write SFM macros for Image SXM (a spin off of NIH Image for SEM, SAM, STM, and SFM) are getting together to write an article on customized image analysis with this program. I have a web site to house the codes. Anyway, maybe a general article like that would pass review, rather than a specific article on one particular macro application. Barry _____________________________________________________________________________ Barry Bickmore "Lisa, just because Dept. of Geological Sciences I don't care doesn't mean Virginia Tech I don't understand!" Blacksburg, VA 24061 -Homer J. Simpson bbickmor@vt.edu (540)231-8575 FAX:(540)231-3386 Barry's preprints/reprints page: http://www.geocities.com/Athens/Parthenon/2671/preprint.html _____________________________________________________________________________ess ------------------------------ Date: Thu, 07 Oct 1999 12:10:49 -0700 From: "Harvey J. Karten" To: nih-image@io.ece.drexel.edu Subject: Mac vs. PC NIH-Image and IMageJ Message-ID: <37FCF039.43AE3962@ucsd.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit As a long time user of NIH-Image, I now find that ImageJ has some really powerful funtions that can't be replicated on NIH-Image. These include the extensive capabilities for dealing with 16 bit multichannel, multi-LUT (8 bit imposed LUT on the 16 bit images), that are really superb. There are constraints in the Java INterface that are less smooth thatn NIH-Image. I hope that those things get ironed out. In our working with ImageJ, we find that it is very robust on both Mac and PC. Even if occasional problems come up, it never crashes the whole damned system the way the bloody Microsoft programs do. Are you sure that you are running the correct versions of Java for each platform, QuickTime, etc. Hats off to Wayne. He again is going against the mainstream, and proving himself to be correct. Get on board and help us develop the software for the 21st Century. regards, Harvey Karten UCSD > Subject: RE: Mac vs. PC > Date: Wed, 06 Oct 1999 16:26:35 -0700 > From: Michael Cammer > To: nih-image@io.ece.drexel.edu > > The fact is that Java is not stable or cross-platform compatible. > Rgeardless of the hype. > NIH-Image is a truly awesome program, but Image J is not there yet. > > ******************************************************************** > * Michael Cammer * Analytical Imaging Facility * Albert Einstein * > * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * > * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * > * personal URL: http://cammer.home.mindspring.com/ * > ******************************************************************** > ------------------------------------------------------------------------ > > Subject: RE: Mac vs. PC > Date: Thu, 7 Oct 1999 10:11:53 +0200 > From: Beat Ludin > To: nih-image@io.ece.drexel.edu > > >The fact is that Java is not stable or cross-platform compatible. > >Rgeardless of the hype. > > What's your definition of stability and compatibility? I'm using a > pretty big piece of telebanking software written entirely in Java. > Runs flawlessly under MacOS, Win95/98/NT and, from what I hear, > various UNIX flavors. Tested ImageJ under MacOS and Win98 and it ran > just fine. Both applications have been perfectly stable so far. I > would say Java is as cross- platform compatible and stable as the > virtual machine running it and the underlying OS - and the VMs have > become pretty good lately (make sure you run the latest version > available). Java is a more programmer-friendly language than, for > instance, C++, too. As long as the guys in Redmond are not succeeding > in screwing up the standard with their "embrace&extend" strategy, > Java is probably here to stay and grow. Regardless of the neg-hype :-) > > > NIH-Image is a truly awesome program, but Image J is not there yet. > > Well, ImageJ does a few things already that NIH-Image does not, and > I'm sure Wayne is working day and night to implement the missing > features. > > Just my two dark current electrons > > Beat > > LIFE IMAGING SERVICES > - visit our web site at http://www.lis.ch > +----------------------------------------------------------- > | Dr. B. Ludin > | Life Imaging Services fon ++41 (0)79 235 7154 > | Muehletalweg 22 fax ++41 (0)86 062 296 3160 NEW! > | CH-4600 Olten beat.ludin@lis.ch > | Switzerland http://www.lis.ch > > ------------------------------------------------------------------------ > -- Harvey J. Karten, M.D. Dept. of Neurosciences University of California at San Diego La Jolla, CA 92093 Phone (Voice) 619-534-4938 FAX 619-534-6602 EMail: hjkarten@ucsd.edu Retina Information Home Page URL: http://www-cajal.ucsd.edu Tayana Manual: ftp://ftp.sailnet.com/lists/tayana/t-37manual/ ------------------------------ Date: Thu, 07 Oct 1999 15:10:55 -0500 From: "Michael J. Herron" To: hjkarten@ucsd.edu CC: nih-image@io.ece.drexel.edu Subject: Re: Mac vs. PC NIH-Image and IMageJ Message-Id: <37FCFE4E.DAF0F222@maroon.tc.umn.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I was complaining to Wayne a couple of weeks ago that ImageJ was crashing my machine. I have since found that it works fine if I run it as a standalone applet instead of trying to run it over the web. I concur ImageJ has some great features esp. for confocalers. Is there any way to use ImageJ to grab from a trusty old Scion LG3? "Harvey J. Karten" wrote: > > As a long time user of NIH-Image, I now find that ImageJ has some really powerful > funtions that can't be replicated on NIH-Image. These include the extensive > capabilities for dealing with 16 bit multichannel, multi-LUT (8 bit imposed LUT on > the 16 bit images), that are really superb. There are constraints in the Java > INterface that are less smooth thatn NIH-Image. I hope that those things get ironed > out. In our working with ImageJ, we find that it is very robust on both Mac and PC. > Even if occasional problems come up, it never crashes the whole damned system the > way the bloody Microsoft programs do. Are you sure that you are running the correct > versions of Java for each platform, QuickTime, etc. > > Hats off to Wayne. He again is going against the mainstream, and proving himself to > be correct. Get on board and help us develop the software for the 21st Century. > > regards, Harvey Karten > UCSD > > > Subject: RE: Mac vs. PC > > Date: Wed, 06 Oct 1999 16:26:35 -0700 > > From: Michael Cammer > > To: nih-image@io.ece.drexel.edu > > > > The fact is that Java is not stable or cross-platform compatible. > > Rgeardless of the hype. > > NIH-Image is a truly awesome program, but Image J is not there yet. > > > > ******************************************************************** > > * Michael Cammer * Analytical Imaging Facility * Albert Einstein * > > * College of Medicine * 1300 Morris Park Ave. * Bronx, NY 10461 * > > * (718) 430-2890 * work URL: http://www.ca.aecom.yu.edu/aif/ * > > * personal URL: http://cammer.home.mindspring.com/ * > > ******************************************************************** > > ------------------------------------------------------------------------ > > > > Subject: RE: Mac vs. PC > > Date: Thu, 7 Oct 1999 10:11:53 +0200 > > From: Beat Ludin > > To: nih-image@io.ece.drexel.edu > > > > >The fact is that Java is not stable or cross-platform compatible. > > >Rgeardless of the hype. > > > > What's your definition of stability and compatibility? I'm using a > > pretty big piece of telebanking software written entirely in Java. > > Runs flawlessly under MacOS, Win95/98/NT and, from what I hear, > > various UNIX flavors. Tested ImageJ under MacOS and Win98 and it ran > > just fine. Both applications have been perfectly stable so far. I > > would say Java is as cross- platform compatible and stable as the > > virtual machine running it and the underlying OS - and the VMs have > > become pretty good lately (make sure you run the latest version > > available). Java is a more programmer-friendly language than, for > > instance, C++, too. As long as the guys in Redmond are not succeeding > > in screwing up the standard with their "embrace&extend" strategy, > > Java is probably here to stay and grow. Regardless of the neg-hype :-) > > > > > NIH-Image is a truly awesome program, but Image J is not there yet. > > > > Well, ImageJ does a few things already that NIH-Image does not, and > > I'm sure Wayne is working day and night to implement the missing > > features. > > > > Just my two dark current electrons > > > > Beat > > > > LIFE IMAGING SERVICES > > - visit our web site at http://www.lis.ch > > +----------------------------------------------------------- > > | Dr. B. Ludin > > | Life Imaging Services fon ++41 (0)79 235 7154 > > | Muehletalweg 22 fax ++41 (0)86 062 296 3160 NEW! > > | CH-4600 Olten beat.ludin@lis.ch > > | Switzerland http://www.lis.ch > > > > ------------------------------------------------------------------------ > > > > -- > Harvey J. Karten, M.D. > Dept. of Neurosciences > University of California at San Diego > La Jolla, CA 92093 > Phone (Voice) 619-534-4938 > FAX 619-534-6602 > EMail: hjkarten@ucsd.edu > Retina Information Home Page > URL: http://www-cajal.ucsd.edu > Tayana Manual: ftp://ftp.sailnet.com/lists/tayana/t-37manual/ -- _______________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /__________________________________________/ ------------------------------ Date: Thu, 7 Oct 1999 16:57:26 -0400 From: Wayne Rasband To: nih-image@io.ece.drexel.edu Subject: Re: Mac vs. PC NIH-Image and IMageJ Message-Id: Content-Type: text/plain; charset="us-ascii" >Is there any way to use ImageJ to grab from a trusty old Scion LG3? This is on the ImageJ to-do list, which I have attached. -wayne 1. Particle analysis. 2. Interactive thresholding and density slicing (bilevel thresholds). 3. Keyboard shortcuts for plug-ins. 4. A "mark and count" tool similar to the one in NIH Image. 5. Scripting language similar to NIH Image's Pascal-like macro language. 6. Display measurement results in a spreadsheet-like table with no arbitrary size limits. 7. Plug-ins that use QuickTime for Java to read and write QuickTime movies. 8. Surface plot plug-in. 9. Density calibration using the Simplex fitting method (straight line, polynomial, exponential, power and log fits) used by NIH image. 10. Fourier transforms. 11. On-line help. 12. Scion LG-3 frame grabber acquisition plug-in. 13. User modifiable convolution kernels. ------------------------------ Date: Thu, 7 Oct 1999 19:25:19 -0400 From: g jones To: nih-image@io.ece.drexel.edu Subject: Re: Publishing/publicising macros Message-Id: Content-Type: text/plain; charset="us-ascii" >Date: Thu, 7 Oct 1999 16:39:26 +0100 >From: Dr Gareth Rees >To: nih-image@io.ece.drexel.edu >Subject: Publishing/publicising macros >Message-Id: >Content-Type: text/plain; charset="us-ascii" >MIME-Version: 1.0 > >Hello imagers, > >Advice wanted! > >Partly for my own work (mostly satellite image processing), and sometimes >just for fun, I write quite a few Image macros to do all sorts of image >processing operations (random examples: terrain visualisation; geometric >rectification; majority-filtering; automatic thresholding...). I pass these >on to colleagues if they are interested in them (I quite often develop them >for colleagues in the first place), and I am gradually starting to put them >on a web site, but does anyone out there have other suggestions on how to >make them more widely known? I did try submitting a couple of articles to >journals, but the referees basically took the view that if a particular >functionality is widely available already, even if only in expensive image >processing suites, then there's no point in publishing implementations in >Image. I disagree with this attitude for two reasons: (1) not everyone can >afford the latest version of Erdas Imagine, or whatever; (2) if I can see >how a macro has been put together, there's some chance that I can actually >understand what it does! > >I am not saying that I have some radically wonderful new image processing >algorithms (I don't - but I find them useful myself), but if anyone is >sympathetic to my reasons (1) and (2) and has any bright ideas... > >Gareth Rees > If nothing else, send them to this forum with a sparing description since they'll be archived and searchable... someday even by mere mortals with a friendlier interface instead of just the unix-speak we have now. Those of us who use image are automatically those who don't spend big bucks on software. Others don't need to care and won't see them anyway. Maybe Wayne will even like them and keep them on zippy. ------------------------------ Date: Fri, 08 Oct 1999 11:41:41 +1000 From: Timothy Bates To: Dr Gareth Rees , Subject: Re: Publishing/publicising macros Message-ID: Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Putting them on a web site and then mentioning that from time to time, where appropriate, on lists such as this, newsgroups etc will soon get you widely known. Another good idea is to put a button on the web page to allow people to subscribe to an updates mailing list. A good example is Denis Pelli's video toolbox site. There are now hundreds of users of his procedures and the site is very well known. Once established, a resource such as this might well warrant an article in BRM&C. Another example is the Tarr and Williams RSVP page. These things can grow to foster a community of users. Thus I wrote a support page for RSVP: And now Lawrence D'Oliviero and I are are developing an AppleScriptable derivative of RSVP called PsyScript. Just get the web page up and your audience is approximately 1000,000 times larger than ANY journal can hope to offer you. Plus no one is going to read the paper journal beyond 6months whereas the web page is finger-tip close. best, tim Thereon 10/8/99 1:39 AM, Dr Gareth Rees wrote: > Hello imagers, > Partly for my own work (mostly satellite image processing), and sometimes > just for fun, I write quite a few Image macros to do all sorts of image > processing operations (random examples: terrain visualisation; geometric > rectification; majority-filtering; automatic thresholding...). I pass these > on to colleagues if they are interested in them (I quite often develop them > for colleagues in the first place), and I am gradually starting to put them > on a web site, but does anyone out there have other suggestions on how to > make them more widely known? ------------------------------ Date: Thu, 07 Oct 1999 20:49:50 -0700 From: Pat and Leslie Pringle To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #230 Message-ID: <37FD69D8.C738E002@thurston.com> Content-Type: text/plain; charset=iso-8859-1; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 8bit Hello imagers, re: Gareth Rees's comments, which included the following: "...I am gradually starting to put them (macros) on a web site, but does anyone out there have other suggestions on how to make them more widely known? I did try submitting a couple of articles to journals, but the referees basically took the view that if a particular functionality is widely available already, even if only in expensive image processing suites, then there's no point in publishing implementations in Image..." Well, I would suggest that perhaps certain publishers often have a different agenda than you or I—and they can change that agenda, or those tendencies, from year to year, depending on editors, megatrends, sunspots, and El Nino. Perhaps you should wait it out and submit again later. Then again, what would we researchers do without teamwork and cooperation such as you are describing in your email as your motivation for producing macros--so aiming at different (less widely known) publication media probably is advisable. Should we presuppose that all good science is focused on particular blue chipjournals?—NOT Plenty of people are doing good science with the most minimal of tools. It's people such as you, Wayne, Norbert Vischer, Greg Joss, Wayne Schuette, Ard, Beat, and numerous others who are making this possible. The collective "we" is stupendously grateful for your contributions. Consider that there are plenty of "top" scientists who neglect (or refuse?) even to put their supporting raw data in repositories or on the web even after they've published in blue chip journals. In this case, the :"great science" is more often than not self supporting with a healthy dose of "trust me", and not necessarily constructive to the whole edifice of scientific progress. You do, of course have, to worry about your own security and vitality as a scientist, so publication ("publish or perish") may be a must and may the Force be with you. But my suggestion would be to not underestimate the power of lower wattage journals or mediums (web sites) for much of what you do, while at the same time continuing your search for the right location having greater visibility and prestige. Let us know what the web page is when you're ready for that. Cheers, Pat Pringle -------------------------------- End of nih-image-d Digest V99 Issue #231 **************************************** From nih-image-request@io.ece.drexel.edu Sat Oct 9 03:24 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id DAA00170; Sat, 9 Oct 1999 03:24:35 -0400 (EDT) Resent-Date: Sat, 9 Oct 1999 03:24:35 -0400 (EDT) User-Agent: Microsoft Outlook Express Macintosh Edition - 5.0 (1507) Date: Sat, 09 Oct 1999 17:07:25 +1000 Subject: Re: Mac vs. PC NIH-Image and IMageJ From: Timothy Bates To: Wayne Rasband , Message-ID: In-Reply-To: Mime-version: 1.0 Content-transfer-encoding: 7bit Resent-Message-ID: <"-wQZq3.0.uc6.Kdk_t"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1817 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 624 Status: O on 10/8/99 6:57 AM, Wayne Rasband wrote: > 5. Scripting language similar to NIH Image's Pascal-like macro language. Hi Wayne. You know, of course, that Apple's new Java Interpreter gives you AppleScript for free, and good scriptability for not too much more. That would absolutely rock: AppleScript is dead easy to learn and hard to forget, so many people could use it. tell ImageJ to repeat with n from 1 to (count of files in folder x) set the Image to open file n of folder x filter Image using filter myFilter with parameters {power:2, length:3} save result as (name of Image) & n end repeat end tell tim From nih-image-d-request@io.ece.drexel.edu Sun Oct 10 06:09 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA07434; Sun, 10 Oct 1999 06:09:39 -0400 (EDT) Date: Sun, 10 Oct 1999 06:09:39 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910101009.GAA07434@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #232 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/232 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 1309 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 232 Today's Topics: Re: Mac vs. PC NIH-Image and IMageJ [ Timothy Bates To: Wayne Rasband , Subject: Re: Mac vs. PC NIH-Image and IMageJ Message-ID: Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit on 10/8/99 6:57 AM, Wayne Rasband wrote: > 5. Scripting language similar to NIH Image's Pascal-like macro language. Hi Wayne. You know, of course, that Apple's new Java Interpreter gives you AppleScript for free, and good scriptability for not too much more. That would absolutely rock: AppleScript is dead easy to learn and hard to forget, so many people could use it. tell ImageJ to repeat with n from 1 to (count of files in folder x) set the Image to open file n of folder x filter Image using filter myFilter with parameters {power:2, length:3} save result as (name of Image) & n end repeat end tell tim -------------------------------- End of nih-image-d Digest V99 Issue #232 **************************************** From nih-image-request@io.ece.drexel.edu Mon Oct 11 05:50 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA14133; Mon, 11 Oct 1999 05:50:35 -0400 (EDT) Resent-Date: Mon, 11 Oct 1999 05:50:35 -0400 (EDT) X-Sender: ajonker.public.amc@reddwarf.amc.uva.nl Message-Id: In-Reply-To: <199910061946.PAA01368@io.ece.drexel.edu> Mime-Version: 1.0 Date: Mon, 11 Oct 1999 11:26:24 +0200 To: nih-image@io.ece.drexel.edu From: Ard Jonker Subject: Re: frame capture Resent-Message-ID: <"OBVyA1.0.LX2.OvQ0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1818 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1269 Status: O >1. NIH-Image has support for digitizing from video using any hardware that >has a vdig component. That includes essentially all Mac-specific >digitization hardware. I do not know how this works with firewire. I tried the MotoDV plugin for Radius Digital Video cards (supposedly FireWire) on a b/w G3 with built-in firewire. Although the plugin happily shows up, no signal can be detected and the image window in the plugin remains black. The plugin itself has the capability to 'capture .... frames', so if you could get it to work you could easily write a time lapse macro. The G3 itself worked in conjunction with the camera that I tried in the above setup, as concluded per trial using a demo version of a firewire digital video capture program. A request for a vdig plugin on the firewire developers mailinglist remained unanswered so far, and I don't expect that to change anymore, unfortunately. Ard dr A.Jonker |Academic Medical Centre Dep of Cell Biology and Histology |Meibergdreef 15 Faculty of Medicine |1105 AZ Amsterdam University of Amsterdam |The Netherlands tel +31 20 566 6804 |fax +31 20 697 4156 PGP fingerprint B263 BEAF 48E2 FE6E 2525 B8C3 EBF9 7685 9A1E 8A3A From nih-image-request@io.ece.drexel.edu Mon Oct 11 12:57 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA28413; Mon, 11 Oct 1999 12:57:49 -0400 (EDT) Resent-Date: Mon, 11 Oct 1999 12:57:49 -0400 (EDT) Message-ID: <380212E5.3F291163@ucsd.edu> Date: Mon, 11 Oct 1999 09:40:06 -0700 From: Harvey Karten Organization: UCSD X-Mailer: Mozilla 4.04 [en] (Win95; U) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Applescript and Java References: <199910101008.GAA07233@io.ece.drexel.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"gOIIz2.0.jA6.ICX0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1819 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 1646 Status: O Tim, Nice suggestion for when we work on the Apple platform. But is a similar scripting language (with the same syntax?) available on a PC Java platform? We now work across both platforms, and benefit from the cross platform compatibility of ImageJ. We are presently trying to repair our plugins that only work on PCs (due to the 32 byte limit on file names in the MacOS - this should be easier with OS X, but that won't be until next year.) Harvey Karten nih-image-d-request@io.ece.drexel.edu wrote: > Subject: > > nih-image-d Digest Volume 99 : Issue 232 > > Today's Topics: > Re: Mac vs. PC NIH-Image and IMageJ [ Timothy Bates > ------------------------------------------------------------------------ > > Subject: Re: Mac vs. PC NIH-Image and IMageJ > Date: Sat, 09 Oct 1999 17:07:25 +1000 > From: Timothy Bates > To: Wayne Rasband , > > on 10/8/99 6:57 AM, Wayne Rasband wrote: > > 5. Scripting language similar to NIH Image's Pascal-like macro language. > > Hi Wayne. You know, of course, that Apple's new Java Interpreter gives you > AppleScript for free, and good scriptability for not too much more. That > would absolutely rock: AppleScript is dead easy to learn and hard to forget, > so many people could use it. > > tell ImageJ to > repeat with n from 1 to (count of files in folder x) > set the Image to open file n of folder x > filter Image using filter myFilter with parameters {power:2, length:3} > save result as (name of Image) & n > end repeat > end tell > tim From nih-image-request@io.ece.drexel.edu Mon Oct 11 19:39 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA02113; Mon, 11 Oct 1999 19:39:37 -0400 (EDT) <380212E5.3F291163@ucsd.edu> Resent-Date: Mon, 11 Oct 1999 19:39:37 -0400 (EDT) Mime-Version: 1.0 X-Sender: scm@128.250.6.196 Message-Id: In-Reply-To: <380212E5.3F291163@ucsd.edu> References: <199910101008.GAA07233@io.ece.drexel.edu> <380212E5.3F291163@ucsd.edu> Date: Tue, 12 Oct 1999 09:25:49 +1000 To: nih-image@io.ece.drexel.edu From: Steve Martin Subject: Re: Applescript and Java Resent-Message-ID: <"D9mBN3.0.FI7.C8d0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1820 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" ; format="flowed" Content-Length: 2025 Status: O At 9:40 AM -0700 on 11/10/99, Harvey Karten wrote: > Tim, > Nice suggestion for when we work on the Apple platform. But is a similar > scripting language (with the same syntax?) available on a PC Java >platform? We now > work across both platforms, and benefit from the cross platform >compatibility of > ImageJ. We are presently trying to repair our plugins that only >work on PCs (due to > the 32 byte limit on file names in the MacOS - this should be >easier with OS X, but > that won't be until next year.) Harvey, if the plugins could be accessed within a .jar file, they should work on Mac OS, as the classes would be accessed within the VM, not the Mac OS file system. This might require some changes to the way Image/J accesses the plugins, ...Wayne? Steve Martin > > Harvey Karten > > nih-image-d-request@io.ece.drexel.edu wrote: > >> Subject: >> >> nih-image-d Digest Volume 99 : Issue 232 >> >> Today's Topics: >> Re: Mac vs. PC NIH-Image and IMageJ [ Timothy Bates >>> >> ------------------------------------------------------------------------ >> >> Subject: Re: Mac vs. PC NIH-Image and IMageJ >> Date: Sat, 09 Oct 1999 17:07:25 +1000 >> From: Timothy Bates >> To: Wayne Rasband , >> >> on 10/8/99 6:57 AM, Wayne Rasband wrote: >> > 5. Scripting language similar to NIH Image's Pascal-like macro language. >> >> Hi Wayne. You know, of course, that Apple's new Java Interpreter gives you >> AppleScript for free, and good scriptability for not too much more. That >> would absolutely rock: AppleScript is dead easy to learn and hard to forget, >> so many people could use it. >> >> tell ImageJ to >> repeat with n from 1 to (count of files in folder x) >> set the Image to open file n of folder x >> filter Image using filter myFilter with parameters {power:2, length:3} >> save result as (name of Image) & n >> end repeat >> end tell >> tim From nih-image-d-request@io.ece.drexel.edu Mon Oct 11 22:43 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id WAA02247; Mon, 11 Oct 1999 22:43:55 -0400 (EDT) Date: Mon, 11 Oct 1999 22:43:55 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910120243.WAA02247@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #233 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/233 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 7168 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 233 Today's Topics: Re: frame capture [ Ard Jonker ] Applescript and Java [ Harvey Karten ] Re: Applescript and Java [ Steve Martin To: nih-image@io.ece.drexel.edu Subject: Re: frame capture Message-Id: Content-Type: text/plain; charset="us-ascii" >1. NIH-Image has support for digitizing from video using any hardware that >has a vdig component. That includes essentially all Mac-specific >digitization hardware. I do not know how this works with firewire. I tried the MotoDV plugin for Radius Digital Video cards (supposedly FireWire) on a b/w G3 with built-in firewire. Although the plugin happily shows up, no signal can be detected and the image window in the plugin remains black. The plugin itself has the capability to 'capture .... frames', so if you could get it to work you could easily write a time lapse macro. The G3 itself worked in conjunction with the camera that I tried in the above setup, as concluded per trial using a demo version of a firewire digital video capture program. A request for a vdig plugin on the firewire developers mailinglist remained unanswered so far, and I don't expect that to change anymore, unfortunately. Ard dr A.Jonker |Academic Medical Centre Dep of Cell Biology and Histology |Meibergdreef 15 Faculty of Medicine |1105 AZ Amsterdam University of Amsterdam |The Netherlands tel +31 20 566 6804 |fax +31 20 697 4156 PGP fingerprint B263 BEAF 48E2 FE6E 2525 B8C3 EBF9 7685 9A1E 8A3A ------------------------------ Date: Mon, 11 Oct 1999 09:40:06 -0700 From: Harvey Karten To: nih-image@io.ece.drexel.edu Subject: Applescript and Java Message-ID: <380212E5.3F291163@ucsd.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Tim, Nice suggestion for when we work on the Apple platform. But is a similar scripting language (with the same syntax?) available on a PC Java platform? We now work across both platforms, and benefit from the cross platform compatibility of ImageJ. We are presently trying to repair our plugins that only work on PCs (due to the 32 byte limit on file names in the MacOS - this should be easier with OS X, but that won't be until next year.) Harvey Karten nih-image-d-request@io.ece.drexel.edu wrote: > Subject: > > nih-image-d Digest Volume 99 : Issue 232 > > Today's Topics: > Re: Mac vs. PC NIH-Image and IMageJ [ Timothy Bates > ------------------------------------------------------------------------ > > Subject: Re: Mac vs. PC NIH-Image and IMageJ > Date: Sat, 09 Oct 1999 17:07:25 +1000 > From: Timothy Bates > To: Wayne Rasband , > > on 10/8/99 6:57 AM, Wayne Rasband wrote: > > 5. Scripting language similar to NIH Image's Pascal-like macro language. > > Hi Wayne. You know, of course, that Apple's new Java Interpreter gives you > AppleScript for free, and good scriptability for not too much more. That > would absolutely rock: AppleScript is dead easy to learn and hard to forget, > so many people could use it. > > tell ImageJ to > repeat with n from 1 to (count of files in folder x) > set the Image to open file n of folder x > filter Image using filter myFilter with parameters {power:2, length:3} > save result as (name of Image) & n > end repeat > end tell > tim ------------------------------ Date: Tue, 12 Oct 1999 09:25:49 +1000 From: Steve Martin To: nih-image@io.ece.drexel.edu Subject: Re: Applescript and Java Message-Id: Content-Type: text/plain; charset="us-ascii" ; format="flowed" At 9:40 AM -0700 on 11/10/99, Harvey Karten wrote: > Tim, > Nice suggestion for when we work on the Apple platform. But is a similar > scripting language (with the same syntax?) available on a PC Java >platform? We now > work across both platforms, and benefit from the cross platform >compatibility of > ImageJ. We are presently trying to repair our plugins that only >work on PCs (due to > the 32 byte limit on file names in the MacOS - this should be >easier with OS X, but > that won't be until next year.) Harvey, if the plugins could be accessed within a .jar file, they should work on Mac OS, as the classes would be accessed within the VM, not the Mac OS file system. This might require some changes to the way Image/J accesses the plugins, ...Wayne? Steve Martin > > Harvey Karten > > nih-image-d-request@io.ece.drexel.edu wrote: > >> Subject: >> >> nih-image-d Digest Volume 99 : Issue 232 >> >> Today's Topics: >> Re: Mac vs. PC NIH-Image and IMageJ [ Timothy Bates >>> >> ------------------------------------------------------------------------ >> >> Subject: Re: Mac vs. PC NIH-Image and IMageJ >> Date: Sat, 09 Oct 1999 17:07:25 +1000 >> From: Timothy Bates >> To: Wayne Rasband , >> >> on 10/8/99 6:57 AM, Wayne Rasband wrote: >> > 5. Scripting language similar to NIH Image's Pascal-like macro language. >> >> Hi Wayne. You know, of course, that Apple's new Java Interpreter gives you >> AppleScript for free, and good scriptability for not too much more. That >> would absolutely rock: AppleScript is dead easy to learn and hard to forget, >> so many people could use it. >> >> tell ImageJ to >> repeat with n from 1 to (count of files in folder x) >> set the Image to open file n of folder x >> filter Image using filter myFilter with parameters {power:2, length:3} >> save result as (name of Image) & n >> end repeat >> end tell >> tim ------------------------------ Date: Mon, 11 Oct 1999 19:23:33 PDT From: "Tish Robertson" To: nih-image@io.ece.drexel.edu Subject: where to begin? Message-ID: <19991012022334.98932.qmail@hotmail.com> Content-Type: text/plain; format=flowed Hello, I've been chosen to research the recommended hardware for use with NIH Image. We've got a Power Mac G3 without video input capability. Can someone give me advice on what hardware we need?(We're going to be analyzing zooplankton behavior) and patiently define the following terms, if you can: -video card -s-video input -frame grabber (is it required hardware for analyzing live feed?) Thanks so much for your assistance. -TR ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com -------------------------------- End of nih-image-d Digest V99 Issue #233 **************************************** From nih-image-request@io.ece.drexel.edu Mon Oct 11 22:44 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id WAA02483; Mon, 11 Oct 1999 22:44:53 -0400 (EDT) Resent-Date: Mon, 11 Oct 1999 22:44:53 -0400 (EDT) Message-ID: <19991012022334.98932.qmail@hotmail.com> X-Originating-IP: [209.138.170.92] From: "Tish Robertson" To: nih-image@io.ece.drexel.edu Subject: where to begin? Date: Mon, 11 Oct 1999 19:23:33 PDT Mime-Version: 1.0 Resent-Message-ID: <"OSJDI1.0.g07.Flf0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1821 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; format=flowed Content-Length: 550 Status: O Hello, I've been chosen to research the recommended hardware for use with NIH Image. We've got a Power Mac G3 without video input capability. Can someone give me advice on what hardware we need?(We're going to be analyzing zooplankton behavior) and patiently define the following terms, if you can: -video card -s-video input -frame grabber (is it required hardware for analyzing live feed?) Thanks so much for your assistance. -TR ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Tue Oct 12 07:44 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA01128; Tue, 12 Oct 1999 07:44:06 -0400 (EDT) Resent-Date: Tue, 12 Oct 1999 07:44:06 -0400 (EDT) Mime-Version: 1.0 Message-Id: Date: Tue, 12 Oct 1999 06:28:08 -0500 To: nih-image@io.ece.drexel.edu From: "J.O'Dell" Subject: re: Applescript and Java Resent-Message-ID: <"Fi5801.0.Qp6.Vin0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1822 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1236 Status: O >Tim, > Nice suggestion for when we work on the Apple platform. But is a similar >scripting language (with the same syntax?) available on a PC Java >platform? We now >work across both platforms, and benefit from the cross platform >compatibility of >ImageJ. We are presently trying to repair our plugins that only work on >PCs (due to >the 32 byte limit on file names in the MacOS - this should be easier with >OS X, but >that won't be until next year.) > >Harvey Karten > Hi Harvey, There is a Mac/Windows scripting language called Userland Frontier. On the Mac side, it uses the same Open Scripting Architecture that Applescript does (it actually predates Applescript on the Mac). I have read that it is highly regarded, and there are demos available on the web. Their syntax differs from Applescript, but presumably would be consistent across platforms. As I understand it, the Mac OS Open Scripting Architecture, once implemented in a Mac application, is accessible by other OSA compliant scripting languages. How Frontier would deal with the implementation on the PC side, I don't know, but it seems it would be a good market for them. Hope this helps. Jane O'Dell Dept. of Chem. Engr. University of Wisconsin-Madison From nih-image-request@io.ece.drexel.edu Tue Oct 12 07:46 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA01564; Tue, 12 Oct 1999 07:46:19 -0400 (EDT) Resent-Date: Tue, 12 Oct 1999 07:46:19 -0400 (EDT) Message-ID: <01BF14B6.0E499550@kahoolawe.geo.vu.nl> From: "E.H.van den Berg" To: "'nih-image@io.ece.drexel.edu'" Subject: copying and pasting Date: Tue, 12 Oct 1999 13:31:13 +0200 MIME-Version: 1.0 Resent-Message-ID: <"8k7Jq.0.xz6.Bmn0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1823 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: multipart/mixed; boundary="---- =_NextPart_000_01BF14B6.0E499550" Content-Length: 4668 Status: O ------ =_NextPart_000_01BF14B6.0E499550 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Dear Fellow Imagers, I have been reading the messages in the usergroup with great interest in = the past months. However, after programming a macro in Scion Image (PC = version) I have come across a problem. Would you take a minute and look = at a specific part of the macro where I think something goes wrong, but = I haven't got a solution. What I am trying to do is to remove particles = touching the edge of an image. The macro does the following operations: 1 - it measures all locations (X and Y) of particles not touching the = edge. 2 - selects all grains individually and copies them to an empty image = with the same dimensions. The problem is that the grains have interior holes. Selecting a grain = using the X and Y locations which is close to or in an interior hole = does not select the grain but the interior hole itself. Because this = introduces an error in the total number of grains subsequent = measurements also contain this error. Is there someone who has an idea = to select a grain without the effect of interior holes? Thanks very much in advance, Sincere greetings, Elmer van den Berg Vrije Universiteit Amsterdam=00=00 ------ =_NextPart_000_01BF14B6.0E499550 Content-Type: application/ms-tnef Content-Transfer-Encoding: base64 eJ8+Ig4LAQaQCAAEAAAAAAABAAEAAQeQBgAIAAAA5AQAAAAAAADoAAEIgAcAGAAAAElQTS5NaWNy b3NvZnQgTWFpbC5Ob3RlADEIAQ2ABAACAAAAAgACAAEEkAYA4AEAAAEAAAAQAAAAAwAAMAIAAAAL AA8OAAAAAAIB/w8BAAAAVQAAAAAAAACBKx+kvqMQGZ1uAN0BD1QCAAAAAG5paC1pbWFnZUBpby5l Y2UuZHJleGVsLmVkdQBTTVRQAG5paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdQAAAAAeAAIwAQAA AAUAAABTTVRQAAAAAB4AAzABAAAAHAAAAG5paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdQADABUM AQAAAAMA/g8GAAAAHgABMAEAAAAeAAAAJ25paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdScAAAAC AQswAQAAACEAAABTTVRQOk5JSC1JTUFHRUBJTy5FQ0UuRFJFWEVMLkVEVQAAAAADAAA5AAAAAAsA QDoBAAAAHgD2XwEAAAAcAAAAbmloLWltYWdlQGlvLmVjZS5kcmV4ZWwuZWR1AAIB918BAAAAVQAA AAAAAACBKx+kvqMQGZ1uAN0BD1QCAAAAAG5paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdQBTTVRQ AG5paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdQAAAAADAP1fAQAAAAMA/18AAAAAAgH2DwEAAAAE AAAAAAAAAtlpAQSAAQAUAAAAY29weWluZyBhbmQgcGFzdGluZwBiBwEFgAMADgAAAM8HCgAMAA0A HwANAAIAJwEBIIADAA4AAADPBwoADAAKADEABAACAC0BAQmAAQAhAAAAMkVCNUIxNkE2RDgwRDMx MTk4MTMwMEUwMjkxQjU3MDkA8AYBA5AGAIwGAAAgAAAACwACAAEAAAALACMAAAAAAAMAJgAAAAAA CwApAAAAAAADADYAAAAAAEAAOQCgMRRKpRS/AR4AcAABAAAAFAAAAGNvcHlpbmcgYW5kIHBhc3Rp bmcAAgFxAAEAAAAWAAAAAb8UpUoParG1MIBtEdOYEwDgKRtXCQAAHgAeDAEAAAAFAAAAU01UUAAA AAAeAB8MAQAAAA8AAABiZXJnQGdlby52dS5ubAAAAwAGECYAT2gDAAcQiwMAAB4ACBABAAAAZQAA AERFQVJGRUxMT1dJTUFHRVJTLElIQVZFQkVFTlJFQURJTkdUSEVNRVNTQUdFU0lOVEhFVVNFUkdS T1VQV0lUSEdSRUFUSU5URVJFU1RJTlRIRVBBU1RNT05USFNIT1dFVkVSLEEAAAAAAgEJEAEAAACF AwAAgQMAABEFAABMWkZ1qf4YnQMACgByY3BnMTI1FjIA+Atgbg4QMDMznQH3IAKkA+MCAGNoCsBg c2V0MCAHEwKAfTkKgXVjAFALAwu1IERiZQrBRmVsCQAH4EnlAMBnBJBzLAqiCoQKgIRJIBEAdmUg YgnhEiAJcGFkC4BnIHT2aBWQB4FzFCEEIAuAFnOadREwcgnACGBwIAPwtRaAIAnBYQVAC4B0BJB/ B5AY8hZzCrAZcQRgAjBoUHMuIEgT0GUVgHL4LCBhAYAEkBoAA2AJwOhhbW0WQmEWsADQA2AxF0JT Y2kCIBP0ICi4UEMgGyEAkAIgKRPw/xVUBaAHgBtgHNEEERyQG9EpAmBlbRrAVwhgbGTMIHkIYBZw YWsfcRawbQuAdRkwG2BuIPAJAG/OaxtgBUAckHNwBZAGkMcN4BoBACAgb2YWdBzDfncWkAlwHsEW gAuAIqBzpx9RJUIWYGdvB5F3A2CbDyAbUGIh8B7FbicFQP8mQCLTBvAh8B1hIJERACcisxwgFnBy eRZDHPBkHPH/BCAp8QlwBGAVgSOiDeAgYP8qYhIwJfMWggmAHdEj8QOR9wdwFCEawFQkNyogK5IW kXcCEBOyFkJvIyAcECiCc+I6FIQxIC0XQBpBE1C+cwhwB5EHQAMgCQBjL8R1HfBYIiNZHrAj8Sso busn8SvPLhSEMjChETAgYHxjdDF0HAELgBcyFjB2aGlkdTGReSIjBaBw/wiQLoMpYRzwA5EgcAUw N8D/LUMYVBaCFvAfYRYwB4AAgf8v8TU1FIQtsiAlKkMo8haC3zalFWMZEx1gBcBoBvAHkP8awAZg NgIcVDajF7EWRjKF/zHJJMAN4BiQKlErcB/AFZD/KfEFsRdRLRI+qi5EM+I15L89iCbzFoJELBhw NeFmGsD6QgWQYRfBJTIXMimABHD3EjAxYjkhcgNgQ6MWginw/wGQAyAh4AbQG6Ej8TalMTD6YhEw cQpQAjAw9jsBNjN/JaAfMQIwQHJJA0ozGsBJ/y6DJPElogIgOeE/IBVRSeP/N2ATUCniRZVANhhi CGA9dG8BEUXCI/E+nD87nABwazcEIBshN8BtK+FDs2R29wBwScAUe1MLgEnAJPEJwvM/4hRsRWwH gAXAVyEqEK8V0UiAF/AUhFYFEGoVkJ5VAwAeQxkwMNFBbRlwXQSQZBwgFIQR8QBeAAAAAAMAEBAA AAAAAwAREAAAAAADAIAQ/////0AABzBQQ96ijhS/AUAACDBQQ96ijhS/AQsAAIAIIAYAAAAAAMAA AAAAAABGAAAAAAOFAAAAAAAAAwACgAggBgAAAAAAwAAAAAAAAEYAAAAAEIUAAAAAAAADAAWACCAG AAAAAADAAAAAAAAARgAAAABShQAAtw0AAB4AJYAIIAYAAAAAAMAAAAAAAABGAAAAAFSFAAABAAAA BAAAADguMAADACaACCAGAAAAAADAAAAAAAAARgAAAAABhQAAAAAAAAsAL4AIIAYAAAAAAMAAAAAA AABGAAAAAA6FAAAAAAAAAwAwgAggBgAAAAAAwAAAAAAAAEYAAAAAEYUAAAAAAAADADKACCAGAAAA AADAAAAAAAAARgAAAAAYhQAAAAAAAB4AQYAIIAYAAAAAAMAAAAAAAABGAAAAADaFAAABAAAAAQAA AAAAAAAeAEKACCAGAAAAAADAAAAAAAAARgAAAAA3hQAAAQAAAAEAAAAAAAAAHgBDgAggBgAAAAAA wAAAAAAAAEYAAAAAOIUAAAEAAAABAAAAAAAAAB4APQABAAAAAQAAAAAAAAADAA00/TcAAOSO ------ =_NextPart_000_01BF14B6.0E499550-- From nih-image-request@io.ece.drexel.edu Tue Oct 12 07:50 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA02475; Tue, 12 Oct 1999 07:50:47 -0400 (EDT) Resent-Date: Tue, 12 Oct 1999 07:50:47 -0400 (EDT) Message-Id: <38031C6E.C16C59B9@tc.umn.edu> Date: Tue, 12 Oct 1999 06:36:07 -0500 From: Michael Herron Reply-To: Michael J Herron Organization: U of MN, Medicine, Infectious Diseases X-Mailer: Mozilla 4.7 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: where to begin? References: <19991012022334.98932.qmail@hotmail.com> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"4wnMj.0._47.Ron0u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1824 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 2916 Status: O Tish, I am NO expert, but I have set up a couple systems for microscopic image analysis. Video Card: is a generic name for a card that does something with video. Maybe grab frames, display stuff, or even crunch the data in a frame. S- video input: An input port that would allow you to plugin an S video source (s- video camera or tape player) (S-video iteslf is an enhanced for of video with more scan lines than a normal video signal) frame grabber: a card that has a video input and allows your computer to "grab" frames from a video source (camera, tape player) I would recommend that you look into a system from Scion (http://www.scioncorp.com/). They have nice packages that are fairly cheap, and are used by most NIH Image users. These are framegrabbing systems. There is another class of systems that do not use frame grabbers they are digital systems that instead of passing an (analogue) video signal to a framegrabber in your computer they pass a DIGITAL signal to your computer. They are more expensive, higher resolution, and slower. Two companies that make such systems are:http://www.diaginc.com/links.htm and http://www.optronics.com/ This whole subject is sort or a morass, so I am trying to tell you my choices, there are MANY other choices some may be better some worse. I esp like the Scion black and white systems and you can ge the whole shebang for under $3000. Color sounds nice...but for science B&W is almost always prefered. You may think that you want to use the old video in jack on the back of some of the older macs...but the resolution is probably not what you would want. If you plan on watching zoo at low power (ie 20X or less) you might want to look into a high resolution B&W CCD (digital) system instead of the Scion video (analogue) system. This would give you sharper pictures at the low magnification. I am not familiar with these high resolution digital cameras...but I am SURE someone else on the list could help you there. Mike Tish Robertson wrote: > > Hello, > I've been chosen to research the recommended hardware for use with NIH > Image. We've got a Power Mac G3 without video input capability. Can someone > give me advice on what hardware we need?(We're going to be analyzing > zooplankton behavior) and patiently define the following terms, if you can: > > -video card > -s-video input > -frame grabber (is it required hardware for analyzing live feed?) > > Thanks so much for your assistance. > -TR > > ______________________________________________________ > Get Your Private, Free Email at http://www.hotmail.com -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ From nih-image-request@io.ece.drexel.edu Tue Oct 12 08:51 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA12474; Tue, 12 Oct 1999 08:51:15 -0400 (EDT) Resent-Date: Tue, 12 Oct 1999 08:51:15 -0400 (EDT) Message-Id: In-Reply-To: <01BF14B6.0E499550@kahoolawe.geo.vu.nl> Mime-Version: 1.0 Date: Tue, 12 Oct 1999 07:37:32 -0500 To: nih-image@io.ece.drexel.edu From: Arnout Ruifrok Subject: Re: copying and pasting Resent-Message-ID: <"KuD021.0.UQ2.vjo0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1825 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1633 Status: O >Dear Fellow Imagers, > >I have been reading the messages in the usergroup with great interest in >the past months. However, after programming a macro in Scion Image (PC >version) I have come across a problem. Would you take a minute and look at >a specific part of the macro where I think something goes wrong, but I >haven't got a solution. What I am trying to do is to remove particles >touching the edge of an image. The macro does the following operations: >1 - it measures all locations (X and Y) of particles not touching the edge. >2 - selects all grains individually and copies them to an empty image with >the same dimensions. > >The problem is that the grains have interior holes. Selecting a grain >using the X and Y locations which is close to or in an interior hole does >not select the grain but the interior hole itself. Because this introduces >an error in the total number of grains subsequent measurements also >contain this error. Is there someone who has an idea to select a grain >without the effect of interior holes? > >Thanks very much in advance, > >Sincere greetings, > >Elmer van den Berg >Vrije Universiteit Amsterdam I have done this by making an outline around the whole image, creating a continuous object that includes all edge-touching particles, and then selecting that object and deleting it. The only disadvantage is that also objects 1 pixel away from the edge are deleted, but for my application that was no problem. (You can circumvent that by copying the image on an image at least 2 pxels wider and higher). Not an answer to your question, but maybe a solution to your problem. Arnout From nih-image-request@io.ece.drexel.edu Tue Oct 12 09:32 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA20564; Tue, 12 Oct 1999 09:32:13 -0400 (EDT) Resent-Date: Tue, 12 Oct 1999 09:32:13 -0400 (EDT) Message-Id: <199910121319.OAA07956@holyrood.ed.ac.uk> X-Mailer: Microsoft Outlook Express Macintosh Edition - 4.5 (0410) Date: Tue, 12 Oct 1999 14:19:43 +0100 Subject: Re: copying and pasting From: "Jeremy Brown" To: nih-image@io.ece.drexel.edu Mime-version: 1.0 X-Priority: 3 Content-transfer-encoding: 7bit Resent-Message-ID: <"9xF1q.0.0S4.uLp0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1826 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 3423 Status: O Try Inclusion Counter v1.1, available from the website below. It employs a counting frame which works with Oval or Rectangular ROI. I ran into almost exactly the same problem you describe. Autooutline will jump from the frame to a hole under rare circumstances (1 of 200 images depending on the placement of the ROI. Well not a hole actually in this case but an artificial hole made by the counting frame and a curved object touching it at exactly the same point as the autooutline (x,y). The procedures LeftBug and RightBug avoid this problem by moveing the (x,y) position of Autooutline if the returned value for the ROI height is < than that of the original ROI. However, Inclusion Counter v1.1 does not preserve holes in objects because it uses three analyse particles calls to calculate % Area; then number of particles not touching the left side of the ROI and then the mean area of particles which do not touch any sides of the ROI. You may therefore want to excise the relative procedures: ApplyROI; BUGLeft; BUGRight; DeleteLeft; DeleteRight; and the macro, [1] Set Region of Interest If you want to use the procedures on a PC you will need to do something about the file path calls in ApplyROI and [1] Set Region of Interest Jeremy **************************************** Jeremy Brown Research Fellow Department of Veterinary Clinical Studies Royal (Dick) School of Veterinary Studies Easter Bush Veterinary Centre Easter Bush Roslin Midlothian EH25 9RG Scotland UK http://users.netmatters.co.uk/brownj/HomePage.html **************************************** ---------- >From: Arnout Ruifrok >To: nih-image@io.ece.drexel.edu >Subject: Re: copying and pasting >Date: Tue, Oct 12, 1999, 1:37 pm > >>Dear Fellow Imagers, >> >>I have been reading the messages in the usergroup with great interest in >>the past months. However, after programming a macro in Scion Image (PC >>version) I have come across a problem. Would you take a minute and look at >>a specific part of the macro where I think something goes wrong, but I >>haven't got a solution. What I am trying to do is to remove particles >>touching the edge of an image. The macro does the following operations: >>1 - it measures all locations (X and Y) of particles not touching the edge. >>2 - selects all grains individually and copies them to an empty image with >>the same dimensions. >> >>The problem is that the grains have interior holes. Selecting a grain >>using the X and Y locations which is close to or in an interior hole does >>not select the grain but the interior hole itself. Because this introduces >>an error in the total number of grains subsequent measurements also >>contain this error. Is there someone who has an idea to select a grain >>without the effect of interior holes? >> >>Thanks very much in advance, >> >>Sincere greetings, >> >>Elmer van den Berg >>Vrije Universiteit Amsterdam > > I have done this by making an outline around the whole image, creating a > continuous object that includes all edge-touching particles, and then > selecting that object and deleting it. The only disadvantage is that also > objects 1 pixel away from the edge are deleted, but for my application that > was no problem. (You can circumvent that by copying the image on an image > at least 2 pxels wider and higher). Not an answer to your question, but > maybe a solution to your problem. > > Arnout > > From nih-image-request@io.ece.drexel.edu Tue Oct 12 09:46 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA23023; Tue, 12 Oct 1999 09:46:26 -0400 (EDT) Resent-Date: Tue, 12 Oct 1999 09:46:26 -0400 (EDT) Message-Id: <38033843.A6C2356C@tc.umn.edu> Date: Tue, 12 Oct 1999 08:35:07 -0500 From: Michael Herron Reply-To: Michael J Herron Organization: U of MN, Medicine, Infectious Diseases X-Mailer: Mozilla 4.7 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: copying and pasting References: Content-Transfer-Encoding: 7bit Resent-Message-ID: <"5Mw1l1.0.b25.sXp0u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1827 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 920 Status: O Arnout Ruifrok wrote: > . . . > > I have done this by making an outline around the whole image, creating a > continuous object that includes all edge-touching particles, and then > selecting that object and deleting it. The only disadvantage is that also > objects 1 pixel away from the edge are deleted, but for my application that > was no problem. (You can circumvent that by copying the image on an image > at least 2 pxels wider and higher). Not an answer to your question, but > maybe a solution to your problem. > > Arnout or perhaps eroding the image once or twice before deleting the edge? -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ From nih-image-request@io.ece.drexel.edu Tue Oct 12 12:40 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA23524; Tue, 12 Oct 1999 12:40:32 -0400 (EDT) Resent-Date: Tue, 12 Oct 1999 12:40:32 -0400 (EDT) Subject: Re: frame capture Date: Tue, 12 Oct 99 12:21:27 -0400 x-sender: jlr$biol@qc1.qc.edu x-mailer: Claris Emailer 1.1 From: jared rifkin To: "nih-image users group" Mime-Version: 1.0 Message-ID: <10045E1597C@qc1.qc.edu> Resent-Message-ID: <"t0ZAg2.0.Ry4.f0s0u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1828 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 307 Status: O please know that os8.5 and 8.6 do not automatically install the 'firewire extensions' (you need two: 'enabler' and 'support') to achieve the firewire capability. go to the apple web site and download the software- theeeennnnnnnnnnn>> you'll be able to see what comes into the G3 via the firewire port. From nih-image-request@io.ece.drexel.edu Tue Oct 12 13:56 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA06592; Tue, 12 Oct 1999 13:56:23 -0400 (EDT) Resent-Date: Tue, 12 Oct 1999 13:56:23 -0400 (EDT) Date: Tue, 12 Oct 1999 13:41:13 -0500 From: John Elsworth Subject: Quantifying nerve fiber "density" To: nih-image@io.ece.drexel.edu Reply-to: John.Elsworth@Yale.edu Message-id: <380380BA.135A3208@Yale.edu> Organization: Yale University MIME-version: 1.0 X-Mailer: Mozilla 4.06 (Macintosh; I; 68K) Content-transfer-encoding: 7bit References: <199910120243.WAA02074@io.ece.drexel.edu> Resent-Message-ID: <"HDV7m2.0.2w.CBt0u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1829 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Length: 2123 Status: O Greetings, I am attempting to estimate the "density" of immunostained nerve fibers in 50 um sections of brain tissue. I have started to try some ways of doing this and want to see if someone knows a better way or if there are any suggestions with regards to what we are doing. 1) The brain sections are 50 um and are immunostained for tyrosine hydroxylase. There is no counterstain and there is very little background. 2) I am capturing an image of one of the regions of interest and get a "high contrast", black & white, image using the "threshold" effect. 3) I then measure the pixels (black) and calculate the amount of black pixels in the total area to give me an estimate of the fiber density. 4) I then move on to other areas of interest leaving the threshold level and light intensity on the 'scope (Olympus BH-2) set as they were. I also look at other slides of other tissues run in the same immunostaining batch as the original slide. Empirically, it seems to work. What is heavily stained to the eye gives a higer density (greater % area) value by this protocol. I see the following limitations. Because I am measuring at 40 x magnification using a 50 um thick section, I try to focus in the middle of the section, but the slightly out of focus layers above and below this focal plane contribute some to the overall density. I see no way around this problem, but think it is likely consistent through out all samples. A second problem may involve looking at different tissue sections from the same, as well as different animals using the original settings. I not sure, but think that since the sections were stained together that there should be some consistency. Also, I am comparing 2 treatment groups with several animals in each group and any variance should be spread through both treatments relatively evenly. Well, any ideas? TIA ___________________________________ John D. Elsworth Ph.D. Departments Psychiatry and Pharmacology Yale University School of Medicine P.O. Box 208066 333 Cedar Street New Haven, CT 06520 Phone: 203-785-6768 Fax: 203-785-7670 john.elsworth@yale.edu From nih-image-d-request@io.ece.drexel.edu Wed Oct 13 01:31 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id BAA19948; Wed, 13 Oct 1999 01:31:13 -0400 (EDT) Date: Wed, 13 Oct 1999 01:31:13 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910130531.BAA19948@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #234 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/234 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 22799 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 234 Today's Topics: re: Applescript and Java [ "J.O'Dell" ] Re: where to begin? [ Michael Herron ] Quantifying nerve fiber "density" [ John Elsworth To: nih-image@io.ece.drexel.edu Subject: re: Applescript and Java Message-Id: Content-Type: text/plain; charset="us-ascii" >Tim, > Nice suggestion for when we work on the Apple platform. But is a similar >scripting language (with the same syntax?) available on a PC Java >platform? We now >work across both platforms, and benefit from the cross platform >compatibility of >ImageJ. We are presently trying to repair our plugins that only work on >PCs (due to >the 32 byte limit on file names in the MacOS - this should be easier with >OS X, but >that won't be until next year.) > >Harvey Karten > Hi Harvey, There is a Mac/Windows scripting language called Userland Frontier. On the Mac side, it uses the same Open Scripting Architecture that Applescript does (it actually predates Applescript on the Mac). I have read that it is highly regarded, and there are demos available on the web. Their syntax differs from Applescript, but presumably would be consistent across platforms. As I understand it, the Mac OS Open Scripting Architecture, once implemented in a Mac application, is accessible by other OSA compliant scripting languages. How Frontier would deal with the implementation on the PC side, I don't know, but it seems it would be a good market for them. Hope this helps. Jane O'Dell Dept. of Chem. Engr. University of Wisconsin-Madison ------------------------------ Date: Tue, 12 Oct 1999 13:31:13 +0200 From: "E.H.van den Berg" To: "'nih-image@io.ece.drexel.edu'" Subject: copying and pasting Message-ID: <01BF14B6.0E499550@kahoolawe.geo.vu.nl> Content-Type: multipart/mixed; boundary="---- =_NextPart_000_01BF14B6.0E499550" ------ =_NextPart_000_01BF14B6.0E499550 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Dear Fellow Imagers, I have been reading the messages in the usergroup with great interest in = the past months. However, after programming a macro in Scion Image (PC = version) I have come across a problem. Would you take a minute and look = at a specific part of the macro where I think something goes wrong, but = I haven't got a solution. What I am trying to do is to remove particles = touching the edge of an image. The macro does the following operations: 1 - it measures all locations (X and Y) of particles not touching the = edge. 2 - selects all grains individually and copies them to an empty image = with the same dimensions. The problem is that the grains have interior holes. Selecting a grain = using the X and Y locations which is close to or in an interior hole = does not select the grain but the interior hole itself. Because this = introduces an error in the total number of grains subsequent = measurements also contain this error. Is there someone who has an idea = to select a grain without the effect of interior holes? Thanks very much in advance, Sincere greetings, Elmer van den Berg Vrije Universiteit Amsterdam=00=00 ------ =_NextPart_000_01BF14B6.0E499550 Content-Type: application/ms-tnef Content-Transfer-Encoding: base64 eJ8+Ig4LAQaQCAAEAAAAAAABAAEAAQeQBgAIAAAA5AQAAAAAAADoAAEIgAcAGAAAAElQTS5NaWNy b3NvZnQgTWFpbC5Ob3RlADEIAQ2ABAACAAAAAgACAAEEkAYA4AEAAAEAAAAQAAAAAwAAMAIAAAAL AA8OAAAAAAIB/w8BAAAAVQAAAAAAAACBKx+kvqMQGZ1uAN0BD1QCAAAAAG5paC1pbWFnZUBpby5l Y2UuZHJleGVsLmVkdQBTTVRQAG5paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdQAAAAAeAAIwAQAA AAUAAABTTVRQAAAAAB4AAzABAAAAHAAAAG5paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdQADABUM AQAAAAMA/g8GAAAAHgABMAEAAAAeAAAAJ25paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdScAAAAC AQswAQAAACEAAABTTVRQOk5JSC1JTUFHRUBJTy5FQ0UuRFJFWEVMLkVEVQAAAAADAAA5AAAAAAsA QDoBAAAAHgD2XwEAAAAcAAAAbmloLWltYWdlQGlvLmVjZS5kcmV4ZWwuZWR1AAIB918BAAAAVQAA AAAAAACBKx+kvqMQGZ1uAN0BD1QCAAAAAG5paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdQBTTVRQ AG5paC1pbWFnZUBpby5lY2UuZHJleGVsLmVkdQAAAAADAP1fAQAAAAMA/18AAAAAAgH2DwEAAAAE AAAAAAAAAtlpAQSAAQAUAAAAY29weWluZyBhbmQgcGFzdGluZwBiBwEFgAMADgAAAM8HCgAMAA0A HwANAAIAJwEBIIADAA4AAADPBwoADAAKADEABAACAC0BAQmAAQAhAAAAMkVCNUIxNkE2RDgwRDMx MTk4MTMwMEUwMjkxQjU3MDkA8AYBA5AGAIwGAAAgAAAACwACAAEAAAALACMAAAAAAAMAJgAAAAAA CwApAAAAAAADADYAAAAAAEAAOQCgMRRKpRS/AR4AcAABAAAAFAAAAGNvcHlpbmcgYW5kIHBhc3Rp bmcAAgFxAAEAAAAWAAAAAb8UpUoParG1MIBtEdOYEwDgKRtXCQAAHgAeDAEAAAAFAAAAU01UUAAA AAAeAB8MAQAAAA8AAABiZXJnQGdlby52dS5ubAAAAwAGECYAT2gDAAcQiwMAAB4ACBABAAAAZQAA AERFQVJGRUxMT1dJTUFHRVJTLElIQVZFQkVFTlJFQURJTkdUSEVNRVNTQUdFU0lOVEhFVVNFUkdS T1VQV0lUSEdSRUFUSU5URVJFU1RJTlRIRVBBU1RNT05USFNIT1dFVkVSLEEAAAAAAgEJEAEAAACF AwAAgQMAABEFAABMWkZ1qf4YnQMACgByY3BnMTI1FjIA+Atgbg4QMDMznQH3IAKkA+MCAGNoCsBg c2V0MCAHEwKAfTkKgXVjAFALAwu1IERiZQrBRmVsCQAH4EnlAMBnBJBzLAqiCoQKgIRJIBEAdmUg YgnhEiAJcGFkC4BnIHT2aBWQB4FzFCEEIAuAFnOadREwcgnACGBwIAPwtRaAIAnBYQVAC4B0BJB/ B5AY8hZzCrAZcQRgAjBoUHMuIEgT0GUVgHL4LCBhAYAEkBoAA2AJwOhhbW0WQmEWsADQA2AxF0JT Y2kCIBP0ICi4UEMgGyEAkAIgKRPw/xVUBaAHgBtgHNEEERyQG9EpAmBlbRrAVwhgbGTMIHkIYBZw YWsfcRawbQuAdRkwG2BuIPAJAG/OaxtgBUAckHNwBZAGkMcN4BoBACAgb2YWdBzDfncWkAlwHsEW gAuAIqBzpx9RJUIWYGdvB5F3A2CbDyAbUGIh8B7FbicFQP8mQCLTBvAh8B1hIJERACcisxwgFnBy eRZDHPBkHPH/BCAp8QlwBGAVgSOiDeAgYP8qYhIwJfMWggmAHdEj8QOR9wdwFCEawFQkNyogK5IW kXcCEBOyFkJvIyAcECiCc+I6FIQxIC0XQBpBE1C+cwhwB5EHQAMgCQBjL8R1HfBYIiNZHrAj8Sso busn8SvPLhSEMjChETAgYHxjdDF0HAELgBcyFjB2aGlkdTGReSIjBaBw/wiQLoMpYRzwA5EgcAUw N8D/LUMYVBaCFvAfYRYwB4AAgf8v8TU1FIQtsiAlKkMo8haC3zalFWMZEx1gBcBoBvAHkP8awAZg NgIcVDajF7EWRjKF/zHJJMAN4BiQKlErcB/AFZD/KfEFsRdRLRI+qi5EM+I15L89iCbzFoJELBhw NeFmGsD6QgWQYRfBJTIXMimABHD3EjAxYjkhcgNgQ6MWginw/wGQAyAh4AbQG6Ej8TalMTD6YhEw cQpQAjAw9jsBNjN/JaAfMQIwQHJJA0ozGsBJ/y6DJPElogIgOeE/IBVRSeP/N2ATUCniRZVANhhi CGA9dG8BEUXCI/E+nD87nABwazcEIBshN8BtK+FDs2R29wBwScAUe1MLgEnAJPEJwvM/4hRsRWwH gAXAVyEqEK8V0UiAF/AUhFYFEGoVkJ5VAwAeQxkwMNFBbRlwXQSQZBwgFIQR8QBeAAAAAAMAEBAA AAAAAwAREAAAAAADAIAQ/////0AABzBQQ96ijhS/AUAACDBQQ96ijhS/AQsAAIAIIAYAAAAAAMAA AAAAAABGAAAAAAOFAAAAAAAAAwACgAggBgAAAAAAwAAAAAAAAEYAAAAAEIUAAAAAAAADAAWACCAG AAAAAADAAAAAAAAARgAAAABShQAAtw0AAB4AJYAIIAYAAAAAAMAAAAAAAABGAAAAAFSFAAABAAAA BAAAADguMAADACaACCAGAAAAAADAAAAAAAAARgAAAAABhQAAAAAAAAsAL4AIIAYAAAAAAMAAAAAA AABGAAAAAA6FAAAAAAAAAwAwgAggBgAAAAAAwAAAAAAAAEYAAAAAEYUAAAAAAAADADKACCAGAAAA AADAAAAAAAAARgAAAAAYhQAAAAAAAB4AQYAIIAYAAAAAAMAAAAAAAABGAAAAADaFAAABAAAAAQAA AAAAAAAeAEKACCAGAAAAAADAAAAAAAAARgAAAAA3hQAAAQAAAAEAAAAAAAAAHgBDgAggBgAAAAAA wAAAAAAAAEYAAAAAOIUAAAEAAAABAAAAAAAAAB4APQABAAAAAQAAAAAAAAADAA00/TcAAOSO ------ =_NextPart_000_01BF14B6.0E499550-- ------------------------------ Date: Tue, 12 Oct 1999 06:36:07 -0500 From: Michael Herron To: nih-image@io.ece.drexel.edu Subject: Re: where to begin? Message-Id: <38031C6E.C16C59B9@tc.umn.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Tish, I am NO expert, but I have set up a couple systems for microscopic image analysis. Video Card: is a generic name for a card that does something with video. Maybe grab frames, display stuff, or even crunch the data in a frame. S- video input: An input port that would allow you to plugin an S video source (s- video camera or tape player) (S-video iteslf is an enhanced for of video with more scan lines than a normal video signal) frame grabber: a card that has a video input and allows your computer to "grab" frames from a video source (camera, tape player) I would recommend that you look into a system from Scion (http://www.scioncorp.com/). They have nice packages that are fairly cheap, and are used by most NIH Image users. These are framegrabbing systems. There is another class of systems that do not use frame grabbers they are digital systems that instead of passing an (analogue) video signal to a framegrabber in your computer they pass a DIGITAL signal to your computer. They are more expensive, higher resolution, and slower. Two companies that make such systems are:http://www.diaginc.com/links.htm and http://www.optronics.com/ This whole subject is sort or a morass, so I am trying to tell you my choices, there are MANY other choices some may be better some worse. I esp like the Scion black and white systems and you can ge the whole shebang for under $3000. Color sounds nice...but for science B&W is almost always prefered. You may think that you want to use the old video in jack on the back of some of the older macs...but the resolution is probably not what you would want. If you plan on watching zoo at low power (ie 20X or less) you might want to look into a high resolution B&W CCD (digital) system instead of the Scion video (analogue) system. This would give you sharper pictures at the low magnification. I am not familiar with these high resolution digital cameras...but I am SURE someone else on the list could help you there. Mike Tish Robertson wrote: > > Hello, > I've been chosen to research the recommended hardware for use with NIH > Image. We've got a Power Mac G3 without video input capability. Can someone > give me advice on what hardware we need?(We're going to be analyzing > zooplankton behavior) and patiently define the following terms, if you can: > > -video card > -s-video input > -frame grabber (is it required hardware for analyzing live feed?) > > Thanks so much for your assistance. > -TR > > ______________________________________________________ > Get Your Private, Free Email at http://www.hotmail.com -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ ------------------------------ Date: Tue, 12 Oct 1999 07:37:32 -0500 From: Arnout Ruifrok To: nih-image@io.ece.drexel.edu Subject: Re: copying and pasting Message-Id: Content-Type: text/plain; charset="us-ascii" >Dear Fellow Imagers, > >I have been reading the messages in the usergroup with great interest in >the past months. However, after programming a macro in Scion Image (PC >version) I have come across a problem. Would you take a minute and look at >a specific part of the macro where I think something goes wrong, but I >haven't got a solution. What I am trying to do is to remove particles >touching the edge of an image. The macro does the following operations: >1 - it measures all locations (X and Y) of particles not touching the edge. >2 - selects all grains individually and copies them to an empty image with >the same dimensions. > >The problem is that the grains have interior holes. Selecting a grain >using the X and Y locations which is close to or in an interior hole does >not select the grain but the interior hole itself. Because this introduces >an error in the total number of grains subsequent measurements also >contain this error. Is there someone who has an idea to select a grain >without the effect of interior holes? > >Thanks very much in advance, > >Sincere greetings, > >Elmer van den Berg >Vrije Universiteit Amsterdam I have done this by making an outline around the whole image, creating a continuous object that includes all edge-touching particles, and then selecting that object and deleting it. The only disadvantage is that also objects 1 pixel away from the edge are deleted, but for my application that was no problem. (You can circumvent that by copying the image on an image at least 2 pxels wider and higher). Not an answer to your question, but maybe a solution to your problem. Arnout ------------------------------ Date: Tue, 12 Oct 1999 14:19:43 +0100 From: "Jeremy Brown" To: nih-image@io.ece.drexel.edu Subject: Re: copying and pasting Message-Id: <199910121319.OAA07956@holyrood.ed.ac.uk> Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Try Inclusion Counter v1.1, available from the website below. It employs a counting frame which works with Oval or Rectangular ROI. I ran into almost exactly the same problem you describe. Autooutline will jump from the frame to a hole under rare circumstances (1 of 200 images depending on the placement of the ROI. Well not a hole actually in this case but an artificial hole made by the counting frame and a curved object touching it at exactly the same point as the autooutline (x,y). The procedures LeftBug and RightBug avoid this problem by moveing the (x,y) position of Autooutline if the returned value for the ROI height is < than that of the original ROI. However, Inclusion Counter v1.1 does not preserve holes in objects because it uses three analyse particles calls to calculate % Area; then number of particles not touching the left side of the ROI and then the mean area of particles which do not touch any sides of the ROI. You may therefore want to excise the relative procedures: ApplyROI; BUGLeft; BUGRight; DeleteLeft; DeleteRight; and the macro, [1] Set Region of Interest If you want to use the procedures on a PC you will need to do something about the file path calls in ApplyROI and [1] Set Region of Interest Jeremy **************************************** Jeremy Brown Research Fellow Department of Veterinary Clinical Studies Royal (Dick) School of Veterinary Studies Easter Bush Veterinary Centre Easter Bush Roslin Midlothian EH25 9RG Scotland UK http://users.netmatters.co.uk/brownj/HomePage.html **************************************** ---------- >From: Arnout Ruifrok >To: nih-image@io.ece.drexel.edu >Subject: Re: copying and pasting >Date: Tue, Oct 12, 1999, 1:37 pm > >>Dear Fellow Imagers, >> >>I have been reading the messages in the usergroup with great interest in >>the past months. However, after programming a macro in Scion Image (PC >>version) I have come across a problem. Would you take a minute and look at >>a specific part of the macro where I think something goes wrong, but I >>haven't got a solution. What I am trying to do is to remove particles >>touching the edge of an image. The macro does the following operations: >>1 - it measures all locations (X and Y) of particles not touching the edge. >>2 - selects all grains individually and copies them to an empty image with >>the same dimensions. >> >>The problem is that the grains have interior holes. Selecting a grain >>using the X and Y locations which is close to or in an interior hole does >>not select the grain but the interior hole itself. Because this introduces >>an error in the total number of grains subsequent measurements also >>contain this error. Is there someone who has an idea to select a grain >>without the effect of interior holes? >> >>Thanks very much in advance, >> >>Sincere greetings, >> >>Elmer van den Berg >>Vrije Universiteit Amsterdam > > I have done this by making an outline around the whole image, creating a > continuous object that includes all edge-touching particles, and then > selecting that object and deleting it. The only disadvantage is that also > objects 1 pixel away from the edge are deleted, but for my application that > was no problem. (You can circumvent that by copying the image on an image > at least 2 pxels wider and higher). Not an answer to your question, but > maybe a solution to your problem. > > Arnout > > ------------------------------ Date: Tue, 12 Oct 1999 08:35:07 -0500 From: Michael Herron To: nih-image@io.ece.drexel.edu Subject: Re: copying and pasting Message-Id: <38033843.A6C2356C@tc.umn.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Arnout Ruifrok wrote: > . . .> > I have done this by making an outline around the whole image, creating a > continuous object that includes all edge-touching particles, and then > selecting that object and deleting it. The only disadvantage is that also > objects 1 pixel away from the edge are deleted, but for my application that > was no problem. (You can circumvent that by copying the image on an image > at least 2 pxels wider and higher). Not an answer to your question, but > maybe a solution to your problem. > > Arnout or perhaps eroding the image once or twice before deleting the edge? -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ ------------------------------ Date: Tue, 12 Oct 99 12:21:27 -0400 From: jared rifkin To: "nih-image users group" Subject: Re: frame capture Message-ID: <10045E1597C@qc1.qc.edu> Content-Type: text/plain; charset="US-ASCII" please know that os8.5 and 8.6 do not automatically install the 'firewire extensions' (you need two: 'enabler' and 'support') to achieve the firewire capability. go to the apple web site and download the software- theeeennnnnnnnnnn>> you'll be able to see what comes into the G3 via the firewire port. ------------------------------ Date: Tue, 12 Oct 1999 13:41:13 -0500 From: John Elsworth To: nih-image@io.ece.drexel.edu Subject: Quantifying nerve fiber "density" Message-id: <380380BA.135A3208@Yale.edu> Content-type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-transfer-encoding: 7bit Greetings, I am attempting to estimate the "density" of immunostained nerve fibers in 50 um sections of brain tissue. I have started to try some ways of doing this and want to see if someone knows a better way or if there are any suggestions with regards to what we are doing. 1) The brain sections are 50 um and are immunostained for tyrosine hydroxylase. There is no counterstain and there is very little background. 2) I am capturing an image of one of the regions of interest and get a "high contrast", black & white, image using the "threshold" effect. 3) I then measure the pixels (black) and calculate the amount of black pixels in the total area to give me an estimate of the fiber density. 4) I then move on to other areas of interest leaving the threshold level and light intensity on the 'scope (Olympus BH-2) set as they were. I also look at other slides of other tissues run in the same immunostaining batch as the original slide. Empirically, it seems to work. What is heavily stained to the eye gives a higer density (greater % area) value by this protocol. I see the following limitations. Because I am measuring at 40 x magnification using a 50 um thick section, I try to focus in the middle of the section, but the slightly out of focus layers above and below this focal plane contribute some to the overall density. I see no way around this problem, but think it is likely consistent through out all samples. A second problem may involve looking at different tissue sections from the same, as well as different animals using the original settings. I not sure, but think that since the sections were stained together that there should be some consistency. Also, I am comparing 2 treatment groups with several animals in each group and any variance should be spread through both treatments relatively evenly. Well, any ideas? TIA ___________________________________ John D. Elsworth Ph.D. Departments Psychiatry and Pharmacology Yale University School of Medicine P.O. Box 208066 333 Cedar Street New Haven, CT 06520 Phone: 203-785-6768 Fax: 203-785-7670 john.elsworth@yale.edu ------------------------------ Date: Wed, 13 Oct 1999 15:19:44 +1000 From: EJ Fjerdingstad To: nih-image@io.ece.drexel.edu Subject: Unidentified subject! Message-Id: Content-Type: text/plain; charset="us-ascii" Dear colleagues, I am having a problem with our image analysis set-up and hope that some may be able to provide me with useful advice. I have a digital camera connected to a dissection microscope, via a video box, and plugged into a Powermac 7500/100. I cannot run NIH image 6.2f on this machine but, when I disconnect the virtual memory, I can run NIH image version 6.1. I wish to measure body parts of ants and thus need to be able to align my specimens accurately and to swivel them around. However, when I do this the image becomes very blurred and moves very jerkily. This has nothing to do with the camera. A colleague uses a different camera in connection with an old PC (a 3/86 I believe) and the program VideoTrace. On their set-up no blurs or jerky movements occur when samples are moved. I attached their camera and phototube to our set-up and now the image was blurred whenever I moved things. When not moving things, the image can be made fairly sharp, though not as good as on their set-up. Now my question, are my problems likely to be related with the fact that we are using an old Mac, or is it more likely to have to do with NIH image handling less images/ time unit than does VideoTrace? If the latter, is the new version of NIH image better? We are considering acquiring one of the new iMacs and should presumably be able to run NIH Image 6.2f on that. Thanks very much for any advice, Else Fjerdingstad ================================== Dr. Else J. Fjerdingstad Postdoctoral researcher Dept. of Genetics and Evolution La Trobe University Bundoora, Victoria 3083 AUSTRALIA Phone office: +61 3 94 79 39 61 Phone lab: +61 3 94 79 50 95 Fax: +61 3 94 79 24 80 Email: fjerding@gen.latrobe.edu.au ================================== "Il faut faire de la vie un reve, et d'un reve faire une realite." Pierre Curie ================================== -------------------------------- End of nih-image-d Digest V99 Issue #234 **************************************** From nih-image-request@io.ece.drexel.edu Wed Oct 13 01:32 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id BAA20148; Wed, 13 Oct 1999 01:32:01 -0400 (EDT) Resent-Date: Wed, 13 Oct 1999 01:32:01 -0400 (EDT) Message-Id: Mime-Version: 1.0 Date: Wed, 13 Oct 1999 15:19:44 +1000 To: nih-image@io.ece.drexel.edu From: EJ Fjerdingstad Resent-Message-ID: <"q3f0c1.0.KD4.yP11u"@io> Resent-From: nih-image@io.ece.drexel.edu Subject: Unidentified subject! Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1830 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1887 Status: O Dear colleagues, I am having a problem with our image analysis set-up and hope that some may be able to provide me with useful advice. I have a digital camera connected to a dissection microscope, via a video box, and plugged into a Powermac 7500/100. I cannot run NIH image 6.2f on this machine but, when I disconnect the virtual memory, I can run NIH image version 6.1. I wish to measure body parts of ants and thus need to be able to align my specimens accurately and to swivel them around. However, when I do this the image becomes very blurred and moves very jerkily. This has nothing to do with the camera. A colleague uses a different camera in connection with an old PC (a 3/86 I believe) and the program VideoTrace. On their set-up no blurs or jerky movements occur when samples are moved. I attached their camera and phototube to our set-up and now the image was blurred whenever I moved things. When not moving things, the image can be made fairly sharp, though not as good as on their set-up. Now my question, are my problems likely to be related with the fact that we are using an old Mac, or is it more likely to have to do with NIH image handling less images/ time unit than does VideoTrace? If the latter, is the new version of NIH image better? We are considering acquiring one of the new iMacs and should presumably be able to run NIH Image 6.2f on that. Thanks very much for any advice, Else Fjerdingstad ================================== Dr. Else J. Fjerdingstad Postdoctoral researcher Dept. of Genetics and Evolution La Trobe University Bundoora, Victoria 3083 AUSTRALIA Phone office: +61 3 94 79 39 61 Phone lab: +61 3 94 79 50 95 Fax: +61 3 94 79 24 80 Email: fjerding@gen.latrobe.edu.au ================================== "Il faut faire de la vie un reve, et d'un reve faire une realite." Pierre Curie ================================== From nih-image-request@io.ece.drexel.edu Wed Oct 13 04:00 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id EAA19200; Wed, 13 Oct 1999 04:00:47 -0400 (EDT) Resent-Date: Wed, 13 Oct 1999 04:00:47 -0400 (EDT) X-Sender: ajonker.public.amc@reddwarf.amc.uva.nl Message-Id: In-Reply-To: <199910130522.BAA17996@io.ece.drexel.edu> Mime-Version: 1.0 Date: Wed, 13 Oct 1999 09:44:41 +0200 To: nih-image@io.ece.drexel.edu From: Ard Jonker Subject: Re: where to begin? Resent-Message-ID: <"OkIFX.0.6v3.0Y31u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1831 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1333 Status: O >If you plan on watching zoo at low power (ie 20X or less) you might want >to look into a high resolution B&W CCD (digital) system instead of the >Scion video (analogue) system. This would give you sharper pictures at >the low magnification. I am not familiar with these high resolution >digital cameras...but I am SURE someone else on the list could help you there. http://www.optronics.com has an interesting camera available (Magnafire) which is 1280x1024x10 bit at 3 colors (filterwheel with dichroic filters) or plain (ie. no filter) with firewire (ie. digital) output and delivers 10 images per second. Download their pdf if you are interested. It should directly connect to a mac, hardware wise. I had contact with the company and they told me that a mac interface is on its way but that may take a while ('It has taken our engineering group more than a year just to create the interface for the PC'). As the Mac OS has firewire support built in, I hope they manage to squeeze this one out a bit faster. I suggested they use the plugin architecture to save them time, and forwarded some appropriate links. Maybe interested parties on this list could show their interest by dropping a mail to mr Dave Mcghee . disclaimer: I have no connections with optronics and haven't seen the camera alive. ard From nih-image-request@io.ece.drexel.edu Wed Oct 13 05:23 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA05349; Wed, 13 Oct 1999 05:23:26 -0400 (EDT) Resent-Date: Wed, 13 Oct 1999 05:23:26 -0400 (EDT) Date: Wed, 13 Oct 1999 05:05:01 -0400 (EDT) From: nih-image-owner@io.ece.drexel.edu Message-Id: <199910130905.FAA01574@io.ece.drexel.edu> To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Resent-Message-ID: <"exduI3.0.nO._i41u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1832 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 3743 Status: O NIH-image Mailing List Help --------------------------- nih-image-d-request@biomed.drexel.edu - Aministration for Digest nih-image-request@biomed.drexel.edu - Administation for List ********************************************************* * DO NOT SEND ADMINISTRATIVE COMMANDS TO THE LIST * ********************************************************* nih-image@biomed.drexel.edu - Sends mail to everyone on the list Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- From nih-image-d-request@io.ece.drexel.edu Thu Oct 14 06:12 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA05722; Thu, 14 Oct 1999 06:12:43 -0400 (EDT) Date: Thu, 14 Oct 1999 06:12:43 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910141012.GAA05722@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #235 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/235 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5966 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 235 Today's Topics: Re: where to begin? [ Ard Jonker ] ADMIN: list commands [ nih-image-owner@io.ece.drexel.edu ] ------------------------------ Date: Wed, 13 Oct 1999 09:44:41 +0200 From: Ard Jonker To: nih-image@io.ece.drexel.edu Subject: Re: where to begin? Message-Id: Content-Type: text/plain; charset="us-ascii" >If you plan on watching zoo at low power (ie 20X or less) you might want >to look into a high resolution B&W CCD (digital) system instead of the >Scion video (analogue) system. This would give you sharper pictures at >the low magnification. I am not familiar with these high resolution >digital cameras...but I am SURE someone else on the list could help you there. http://www.optronics.com has an interesting camera available (Magnafire) which is 1280x1024x10 bit at 3 colors (filterwheel with dichroic filters) or plain (ie. no filter) with firewire (ie. digital) output and delivers 10 images per second. Download their pdf if you are interested. It should directly connect to a mac, hardware wise. I had contact with the company and they told me that a mac interface is on its way but that may take a while ('It has taken our engineering group more than a year just to create the interface for the PC'). As the Mac OS has firewire support built in, I hope they manage to squeeze this one out a bit faster. I suggested they use the plugin architecture to save them time, and forwarded some appropriate links. Maybe interested parties on this list could show their interest by dropping a mail to mr Dave Mcghee . disclaimer: I have no connections with optronics and haven't seen the camera alive. ard ------------------------------ Date: Wed, 13 Oct 1999 05:05:01 -0400 (EDT) From: nih-image-owner@io.ece.drexel.edu To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Message-Id: <199910130905.FAA01574@io.ece.drexel.edu> NIH-image Mailing List Help --------------------------- nih-image-d-request@biomed.drexel.edu - Aministration for Digest nih-image-request@biomed.drexel.edu - Administation for List ********************************************************* * DO NOT SEND ADMINISTRATIVE COMMANDS TO THE LIST * ********************************************************* nih-image@biomed.drexel.edu - Sends mail to everyone on the list Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- -------------------------------- End of nih-image-d Digest V99 Issue #235 **************************************** From nih-image-request@io.ece.drexel.edu Thu Oct 14 09:09 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA12664; Thu, 14 Oct 1999 09:09:00 -0400 (EDT) Resent-Date: Thu, 14 Oct 1999 09:09:00 -0400 (EDT) X-Sender: drmason@mail.interconnect.com.au Message-Id: In-Reply-To: Mime-Version: 1.0 Date: Thu, 14 Oct 1999 12:05:44 +1030 To: nih-image@io.ece.drexel.edu From: Doug Mason Subject: Re: Unidentified subject! Resent-Message-ID: <"y2djw2.0.Ou1.E3T1u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1833 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1013 Status: O Else Fjerdingstad wrote: [snip] > swivel them around. However, when I do this the image >becomes very blurred and moves very jerkily. [snip] > Now my question, are my problems likely to be related with the fact >that we are using an old Mac, or is it more likely to have to do with NIH >image handling less images/ time unit than does VideoTrace? If the latter, >is the new version of NIH image better? We are considering acquiring one of >the new iMacs and should presumably be able to run NIH Image 6.2f on that. [snip] Whenever I rotate my microscope stage, the image on my Mac always becomes fragmented for a second or so until it re-establishes itself clearly. I am using a G3/233MHz, Scion LG3 framegrabber, and both NIH-Image 1.60 and Object-Image 1.62. Cheers, Doug Mason Mason Geoscience Pty Ltd Petrological Services for the Exploration and Mining Industry email : drmason@interconnect.com.au post : PO Box 78, Glenside SA 5065, Australia phone : +61-8-83901507 fax : +61-8-83901194 From nih-image-request@io.ece.drexel.edu Thu Oct 14 09:41 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA19056; Thu, 14 Oct 1999 09:41:58 -0400 (EDT) Resent-Date: Thu, 14 Oct 1999 09:41:58 -0400 (EDT) Message-Id: X-Mailer: Novell GroupWise 5.5.2 Date: Thu, 14 Oct 1999 09:20:21 -0400 From: "Marie Bundy" To: Subject: Re: where to begin? Mime-Version: 1.0 Content-Disposition: inline Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by io.ece.drexel.edu id JAA15733 Resent-Message-ID: <"TYn4D.0.5s3.IbT1u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1834 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=US-ASCII Content-Length: 737 Status: O Hello Tish, In regard to a suitable camera, the Pulinix line of CCDcameras are quite good. Your lens systems will depend on whether you are looking at large or small-scale behaviors and, of course, the size of your beasties. Zooplankton behavioral analyses are part of what I "do for a living". Although I'm not an expert in systems hardware, I can perhaps help you aviod some pitfalls I've encountered with the behavioral analyses. If you'd like, you can contact me off-line, and we can chat. Marie Bundy ********************************************** Marie H. Bundy Academy of Natural Sciences Estuarine Research Center (ANS-ERC) 10545 Mackall Rd. St. Leonard, MD 20685 Phone: 410-586-9710 Fax: 410-586-9705 bundy@acnatsci.org From nih-image-request@io.ece.drexel.edu Thu Oct 14 09:45 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA19796; Thu, 14 Oct 1999 09:45:55 -0400 (EDT) Resent-Date: Thu, 14 Oct 1999 09:45:55 -0400 (EDT) Message-Id: <3805DAF9.8438EA82@zi.biologie.uni-muenchen.de> Date: Thu, 14 Oct 1999 15:30:33 +0200 From: Kota Miura Organization: Zoological Institute, LMU Muenchen X-Mailer: Mozilla 4.51 [en] (WinNT; I) X-Accept-Language: en,ja,de-DE Mime-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: [Q] rs232c triggering References: <199910141000.GAA02780@io.ece.drexel.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"ER-lp1.0.Z14.beT1u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1835 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 824 Status: O Dear Imagers, I need a help on writing a macro. I am using Scionimage and planning to control a laserdisc recorder via serial port (Rs-232c). Task is to capture time-lapse sequence (with Scion VG-5), enhance contrast and record it on a laserdisc (sonyLVR4000ap). The laserdisc needs a triggering for this purpose so I need to include commands within the macro. Commands are hex numbers between 0x01 and 0xEF. One problem is that according to Scion, the commands could only be between 0x20 and 0x7F (I am not quite sure why). I believe there should be someway around to use full lenght between 0x01 and 0xEF. If anyone has a macro to control a peripheral with those commands, please post here or send me that macro so I can extend it for my purpose. Thank you in advance. Kota Miura Zoological Institute, Munich Germany From nih-image-request@io.ece.drexel.edu Thu Oct 14 14:07 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id OAA05115; Thu, 14 Oct 1999 14:07:54 -0400 (EDT) Resent-Date: Thu, 14 Oct 1999 14:07:54 -0400 (EDT) Message-ID: From: "Goh, S M" To: "'nih-image@biomed.drexel.edu'" Subject: Grey Level and Profile Plot Date: Thu, 14 Oct 1999 18:50:56 +0100 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Resent-Message-ID: <"gtaPE1.0.-F.lWX1u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1836 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain Content-Length: 204 Status: O May I know why there is a decimal place in the values of the grey level when a profile is plotted? I thought the grey scale consists of integers from 0-255. Thanks. Aaron Goh Imperial College, London From nih-image-request@io.ece.drexel.edu Thu Oct 14 18:22 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA02190; Thu, 14 Oct 1999 18:22:19 -0400 (EDT) Resent-Date: Thu, 14 Oct 1999 18:22:19 -0400 (EDT) Message-Id: <199910142210.SAA29357@io.ece.drexel.edu> Delivered-To: fixup-nih-image@biomed.drexel.edu@fixme X-Mailer: Microsoft Outlook Express Macintosh Edition - 4.5 (0410) Date: Thu, 14 Oct 1999 17:09:32 -0500 Subject: Scion LG-3 Frame Grabber Image Capture Card For Sale (NUBUS) From: "Michael Higgs" To: nih-image@io.ece.drexel.edu Mime-version: 1.0 X-Priority: 3 Content-transfer-encoding: 7bit Resent-Message-ID: <"yMbDb1.0.uA7.3Jb1u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1837 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 506 Status: O To whom it may concern: We have for sale one Scion LG-3 Frame Grabber (Revision C) Image Capture card for NUBUS equipped Macintosh computers with an additional 16MB of RAM installed. All software and manuals included. Interested parties please send e-mail to: sales@imagesmithsinc.com. First e-mail takes card home. The price is $499.99. You can also e-mail any questions to the above address. Thank you. ImageSmiths Incorporated Michael Higgs 212 Commercial Street Stillwater, MN 55082 651.430.3104 From nih-image-d-request@io.ece.drexel.edu Fri Oct 15 06:19 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA27156; Fri, 15 Oct 1999 06:19:24 -0400 (EDT) Date: Fri, 15 Oct 1999 06:19:24 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910151019.GAA27156@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #236 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/236 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5493 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 236 Today's Topics: Re: Unidentified subject! [ Doug Mason ] [Q] rs232c triggering [ Kota Miura ] Scion LG-3 Frame Grabber Image Captu [ "Michael Higgs" To: nih-image@io.ece.drexel.edu Subject: Re: Unidentified subject! Message-Id: Content-Type: text/plain; charset="us-ascii" Else Fjerdingstad wrote: [snip] > swivel them around. However, when I do this the image >becomes very blurred and moves very jerkily. [snip] > Now my question, are my problems likely to be related with the fact >that we are using an old Mac, or is it more likely to have to do with NIH >image handling less images/ time unit than does VideoTrace? If the latter, >is the new version of NIH image better? We are considering acquiring one of >the new iMacs and should presumably be able to run NIH Image 6.2f on that. [snip] Whenever I rotate my microscope stage, the image on my Mac always becomes fragmented for a second or so until it re-establishes itself clearly. I am using a G3/233MHz, Scion LG3 framegrabber, and both NIH-Image 1.60 and Object-Image 1.62. Cheers, Doug Mason Mason Geoscience Pty Ltd Petrological Services for the Exploration and Mining Industry email : drmason@interconnect.com.au post : PO Box 78, Glenside SA 5065, Australia phone : +61-8-83901507 fax : +61-8-83901194 ------------------------------ Date: Thu, 14 Oct 1999 09:20:21 -0400 From: "Marie Bundy" To: Subject: Re: where to begin? Message-Id: Content-Type: text/plain; charset=US-ASCII Content-Disposition: inline Content-Transfer-Encoding: 8bit Hello Tish, In regard to a suitable camera, the Pulinix line of CCDcameras are quite good. Your lens systems will depend on whether you are looking at large or small-scale behaviors and, of course, the size of your beasties. Zooplankton behavioral analyses are part of what I "do for a living". Although I'm not an expert in systems hardware, I can perhaps help you aviod some pitfalls I've encountered with the behavioral analyses. If you'd like, you can contact me off-line, and we can chat. Marie Bundy ********************************************** Marie H. Bundy Academy of Natural Sciences Estuarine Research Center (ANS-ERC) 10545 Mackall Rd. St. Leonard, MD 20685 Phone: 410-586-9710 Fax: 410-586-9705 bundy@acnatsci.org ------------------------------ Date: Thu, 14 Oct 1999 15:30:33 +0200 From: Kota Miura To: nih-image@io.ece.drexel.edu Subject: [Q] rs232c triggering Message-Id: <3805DAF9.8438EA82@zi.biologie.uni-muenchen.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Dear Imagers, I need a help on writing a macro. I am using Scionimage and planning to control a laserdisc recorder via serial port (Rs-232c). Task is to capture time-lapse sequence (with Scion VG-5), enhance contrast and record it on a laserdisc (sonyLVR4000ap). The laserdisc needs a triggering for this purpose so I need to include commands within the macro. Commands are hex numbers between 0x01 and 0xEF. One problem is that according to Scion, the commands could only be between 0x20 and 0x7F (I am not quite sure why). I believe there should be someway around to use full lenght between 0x01 and 0xEF. If anyone has a macro to control a peripheral with those commands, please post here or send me that macro so I can extend it for my purpose. Thank you in advance. Kota Miura Zoological Institute, Munich Germany ------------------------------ Date: Thu, 14 Oct 1999 18:50:56 +0100 From: "Goh, S M" To: "'nih-image@biomed.drexel.edu'" Subject: Grey Level and Profile Plot Message-ID: Content-Type: text/plain May I know why there is a decimal place in the values of the grey level when a profile is plotted? I thought the grey scale consists of integers from 0-255. Thanks. Aaron Goh Imperial College, London ------------------------------ Date: Thu, 14 Oct 1999 17:09:32 -0500 From: "Michael Higgs" To: nih-image@io.ece.drexel.edu Subject: Scion LG-3 Frame Grabber Image Capture Card For Sale (NUBUS) Message-Id: <199910142210.SAA29357@io.ece.drexel.edu> Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit To whom it may concern: We have for sale one Scion LG-3 Frame Grabber (Revision C) Image Capture card for NUBUS equipped Macintosh computers with an additional 16MB of RAM installed. All software and manuals included. Interested parties please send e-mail to: sales@imagesmithsinc.com. First e-mail takes card home. The price is $499.99. You can also e-mail any questions to the above address. Thank you. ImageSmiths Incorporated Michael Higgs 212 Commercial Street Stillwater, MN 55082 651.430.3104 -------------------------------- End of nih-image-d Digest V99 Issue #236 **************************************** From nih-image-request@io.ece.drexel.edu Fri Oct 15 10:42 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id KAA26819; Fri, 15 Oct 1999 10:42:33 -0400 (EDT) Resent-Date: Fri, 15 Oct 1999 10:42:33 -0400 (EDT) Date: Fri, 15 Oct 1999 09:22:14 -0500 From: Chris Liner Subject: Donation notice To: NIH mail list Message-id: <38073894.59377D06@utulsa.edu> MIME-version: 1.0 X-Mailer: Mozilla 4.61 (Macintosh; I; PPC) Content-transfer-encoding: 7bit X-Accept-Language: en Resent-Message-ID: <"SLl5R2.0.Fe5.8Wp1u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1838 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Length: 1508 Status: O Dear N IHers, I have developed a commercial seismic interpretation system around NIH-Image. The system is called Cubic and consists of macros, external utility applications and documentation for applying NIH-Image to visualize and interpret geophysical data. Cubic includes an external gridding program that makes it useful for many kinds of geophysical data (gravity, mag, conductivity, etc.) as well as subsurface geology (structure maps, isopachs, etc.). We use the system at U. of Tulsa to teach the fundamentals of 3D seismic interpretation, including tracking, depth conversion, and rock property estimation from seismic amplitude. Cubic currently in use by universities in Arkansas, Texas and North Carolina. I am interested in additional donations to a U.S. universities before the end of the year. If anyone on the NIH-Image mail list is interested, please contact me directly via email. Information about Cubic can be found at my web site listed below. Dr. Christopher L. Liner Associate Professor Department of Geoscience University of Tulsa 600 S. College Ave Tulsa, OK 74104-3189 --------------------------- cll@utulsa.edu http://douze.utulsa.edu/~cll/ChrisLiner.html 918-631-3028 [phone] 918-631-2091 [fax] -- Dr. Christopher L. Liner Associate Professor Department of Geoscience University of Tulsa 600 S. College Ave Tulsa, OK 74104-3189 --------------------------- cll@utulsa.edu http://douze.utulsa.edu/~cll/ChrisLiner.html 918-631-3028 [phone] 918-631-2091 [fax] From nih-image-request@io.ece.drexel.edu Fri Oct 15 13:46 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA28912; Fri, 15 Oct 1999 13:46:22 -0400 (EDT) Resent-Date: Fri, 15 Oct 1999 13:46:22 -0400 (EDT) Message-Id: <4.2.0.58.19991015122756.0095e410@students.wisc.edu> X-Sender: kahowell@students.wisc.edu X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.0.58 Date: Fri, 15 Oct 1999 12:28:48 -0500 To: nih-image@io.ece.drexel.edu From: Kimberly A Howell Subject: Power Spectrum Mime-Version: 1.0 Resent-Message-ID: <"-19aw.0.jT6.OKs1u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1839 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii"; format=flowed Content-Length: 715 Status: O Hello Fellow NIH - Imagers, I just have a question for you concerning the power spectrum. I want to integrate over all theta at a particular r to obtain the power spectrum as a function of just r. Does any one know of a particular macro that would allow me to do that? I have been investigating the annular selection macro as a way to choose an annular ROI which would be ideal for what I need to do but that seems to be useful only for filtering/masking. If you could reply directly to me at kahowell@students.wisc.edu that would be great. Thank you for your time. Sincerely, Kim Howell Kimberly Howell, M.S. University of Wisconsin - Madison Department of Medical Physics Phone: (608) 263-9257 From nih-image-d-request@io.ece.drexel.edu Sat Oct 16 06:09 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA17469; Sat, 16 Oct 1999 06:09:08 -0400 (EDT) Date: Sat, 16 Oct 1999 06:09:08 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910161009.GAA17469@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #237 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/237 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 3262 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 237 Today's Topics: Donation notice [ Chris Liner ] Power Spectrum [ Kimberly A Howell To: NIH mail list Subject: Donation notice Message-id: <38073894.59377D06@utulsa.edu> Content-type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-transfer-encoding: 7bit Dear N IHers, I have developed a commercial seismic interpretation system around NIH-Image. The system is called Cubic and consists of macros, external utility applications and documentation for applying NIH-Image to visualize and interpret geophysical data. Cubic includes an external gridding program that makes it useful for many kinds of geophysical data (gravity, mag, conductivity, etc.) as well as subsurface geology (structure maps, isopachs, etc.). We use the system at U. of Tulsa to teach the fundamentals of 3D seismic interpretation, including tracking, depth conversion, and rock property estimation from seismic amplitude. Cubic currently in use by universities in Arkansas, Texas and North Carolina. I am interested in additional donations to a U.S. universities before the end of the year. If anyone on the NIH-Image mail list is interested, please contact me directly via email. Information about Cubic can be found at my web site listed below. Dr. Christopher L. Liner Associate Professor Department of Geoscience University of Tulsa 600 S. College Ave Tulsa, OK 74104-3189 --------------------------- cll@utulsa.edu http://douze.utulsa.edu/~cll/ChrisLiner.html 918-631-3028 [phone] 918-631-2091 [fax] -- Dr. Christopher L. Liner Associate Professor Department of Geoscience University of Tulsa 600 S. College Ave Tulsa, OK 74104-3189 --------------------------- cll@utulsa.edu http://douze.utulsa.edu/~cll/ChrisLiner.html 918-631-3028 [phone] 918-631-2091 [fax] ------------------------------ Date: Fri, 15 Oct 1999 12:28:48 -0500 From: Kimberly A Howell To: nih-image@io.ece.drexel.edu Subject: Power Spectrum Message-Id: <4.2.0.58.19991015122756.0095e410@students.wisc.edu> Content-Type: text/plain; charset="us-ascii"; format=flowed Hello Fellow NIH - Imagers, I just have a question for you concerning the power spectrum. I want to integrate over all theta at a particular r to obtain the power spectrum as a function of just r. Does any one know of a particular macro that would allow me to do that? I have been investigating the annular selection macro as a way to choose an annular ROI which would be ideal for what I need to do but that seems to be useful only for filtering/masking. If you could reply directly to me at kahowell@students.wisc.edu that would be great. Thank you for your time. Sincerely, Kim Howell Kimberly Howell, M.S. University of Wisconsin - Madison Department of Medical Physics Phone: (608) 263-9257 -------------------------------- End of nih-image-d Digest V99 Issue #237 **************************************** From nih-image-request@io.ece.drexel.edu Sun Oct 17 08:18 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA00919; Sun, 17 Oct 1999 08:18:12 -0400 (EDT) Resent-Date: Sun, 17 Oct 1999 08:18:12 -0400 (EDT) Message-ID: <3809ADF5.A249EA9E@netvision.net.il> Date: Sun, 17 Oct 1999 14:07:36 +0300 From: asaf kaplan X-Mailer: Mozilla 4.5 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image mailing list Subject: macro needed Content-Transfer-Encoding: 7bit Resent-Message-ID: <"RBXQN1.0.Fo6.qjR2u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1840 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Length: 562 Status: O Hello imagers, I'm a zoology student working with image 1.61 on a power pc mac. I have somewhere around 100 large stacks (about 65-70 slices each) that I need to flip horizontally. Currently I'm doing it manually, working with the flip horizontally and the next slice commands, and as you know it takes its time. Unfortunately I don't know how to make a macro to do it for me (never learned programing). Does someone knows about such macro or can help me write one? Thanks, Asaf Asaf Kaplan Zoology department Tel Aviv University From nih-image-request@io.ece.drexel.edu Sun Oct 17 09:04 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA11483; Sun, 17 Oct 1999 09:04:06 -0400 (EDT) Resent-Date: Sun, 17 Oct 1999 09:04:06 -0400 (EDT) X-Sender: ad1054@unix.worldpath.net Message-Id: Mime-Version: 1.0 Date: Sun, 17 Oct 1999 08:52:04 -0400 To: nih-image@io.ece.drexel.edu From: ad1054@worldpath.net (Dr. Christopher Coulon) Subject: Re: macro needed Resent-Message-ID: <"B7xiF.0.E72.JQS2u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1841 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1712 Status: O >Hello imagers, >I'm a zoology student working with image 1.61 on a power pc mac. I have >somewhere around 100 large stacks (about 65-70 slices each) that I need >to flip horizontally. Currently I'm doing it manually, working with the >flip horizontally and the next slice commands, and as you know it takes >its time. Unfortunately I don't know how to make a macro to do it for me >(never learned programing). > >Does someone knows about such macro or can help me write one? Asaf, You should get into writing macros if you have lots of repetitious tasks -- they are fun to write, especially when they save you from hours of tedious drudgery. The macro you need goes like this: ******* cut and paste the text between these lines ******* macro '[f] flip stack slices'; var j:integer begin selectAll; for j:=1 to nSlices do begin selectSlice(j); FlipHorizontal; killROI; end; end; ******* cut and paste the text between these lines ******* In NIH-Image, choose "new" under the "File" menu and click on the "Text" button. Cut and paste the above macro into the resulting text window. Save it, then do a <9> (hold down the Apple Key and press the "9" key) to load your macro. Make sure the stack you want to rotate is the only image open. Select the stack and press the "f" key. If you need or want help learning the macro language further, get back to me offline. Chris * * * * * * * * * * * * * * * * * * * Dr. Christopher Hunt Coulon * * The GAIA Group * * 40 Chick Road * * Wolfeboro, New Hampshire 03894 * * 603 569-0028 * * * * * * * * * * * * * * * * * * * http://www.worldpath.net/~ad1054/ From nih-image-d-request@io.ece.drexel.edu Mon Oct 18 00:57 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id AAA24037; Mon, 18 Oct 1999 00:57:35 -0400 (EDT) Date: Mon, 18 Oct 1999 00:57:35 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910180457.AAA24037@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #238 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/238 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5719 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 238 Today's Topics: macro needed [ asaf kaplan To: nih-image mailing list Subject: macro needed Message-ID: <3809ADF5.A249EA9E@netvision.net.il> Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit Hello imagers, I'm a zoology student working with image 1.61 on a power pc mac. I have somewhere around 100 large stacks (about 65-70 slices each) that I need to flip horizontally. Currently I'm doing it manually, working with the flip horizontally and the next slice commands, and as you know it takes its time. Unfortunately I don't know how to make a macro to do it for me (never learned programing). Does someone knows about such macro or can help me write one? Thanks, Asaf Asaf Kaplan Zoology department Tel Aviv University ------------------------------ Date: Sun, 17 Oct 1999 08:52:04 -0400 From: ad1054@worldpath.net (Dr. Christopher Coulon) To: nih-image@io.ece.drexel.edu Subject: Re: macro needed Message-Id: Content-Type: text/plain; charset="us-ascii" >Hello imagers, >I'm a zoology student working with image 1.61 on a power pc mac. I have >somewhere around 100 large stacks (about 65-70 slices each) that I need >to flip horizontally. Currently I'm doing it manually, working with the >flip horizontally and the next slice commands, and as you know it takes >its time. Unfortunately I don't know how to make a macro to do it for me >(never learned programing). > >Does someone knows about such macro or can help me write one? Asaf, You should get into writing macros if you have lots of repetitious tasks -- they are fun to write, especially when they save you from hours of tedious drudgery. The macro you need goes like this: ******* cut and paste the text between these lines ******* macro '[f] flip stack slices'; var j:integer begin selectAll; for j:=1 to nSlices do begin selectSlice(j); FlipHorizontal; killROI; end; end; ******* cut and paste the text between these lines ******* In NIH-Image, choose "new" under the "File" menu and click on the "Text" button. Cut and paste the above macro into the resulting text window. Save it, then do a <9> (hold down the Apple Key and press the "9" key) to load your macro. Make sure the stack you want to rotate is the only image open. Select the stack and press the "f" key. If you need or want help learning the macro language further, get back to me offline. Chris * * * * * * * * * * * * * * * * * * * Dr. Christopher Hunt Coulon * * The GAIA Group * * 40 Chick Road * * Wolfeboro, New Hampshire 03894 * * 603 569-0028 * * * * * * * * * * * * * * * * * * * http://www.worldpath.net/~ad1054/ ------------------------------ Date: Mon, 18 Oct 1999 14:49:54 +1000 From: GJOSS@rna.bio.mq.edu.au To: kahowell@students.wisc.edu, nih-image@io.ece.drexel.edu Subject: Re: Power Spectrum Message-ID: <1E2EDF2205A@rna.bio.mq.edu.au> >Date: Fri, 15 Oct 1999 12:28:48 -0500 >To: nih-image@io.ece.drexel.edu >From: Kimberly A Howell >Subject: Power Spectrum > >Hello Fellow NIH - Imagers, > > I just have a question for you concerning the power spectrum. > > I want to integrate over all theta at a particular r to obtain the >power spectrum as a function of just r. Does any one know of a >particular macro that would allow me to do that? I have been >investigating the annular selection macro as a way to choose an annular >ROI which would be ideal for what I need to do but that seems to be >useful only for filtering/masking. > > If you could reply directly to me at kahowell@students.wisc.edu that >would be great. Thank you for your time. > >Sincerely, > Kim Howell > >Kimberly Howell, M.S. >University of Wisconsin - Madison >Department of Medical Physics >Phone: (608) 263-9257 > I expect the following does something like what you want? macro '[R]adial Mean'; var w,h,xc,yc,r,rm,d,pi2,c,b,bins:integer; { Greg Joss 1999-10-18} begin GetPicSize(w,h); pi2:=arctan(1)*8; setUser1Label('dArea'); setUser2Label('dIntens'); setOptions('Area Mean User1 User2'); setPrecision(0,5); xc:=w/2; yc:=h/2; rm:=xc; bins:=getNumber('# bins',rm); resetCounter;setCounter(1); if get('maxMeasurements')',d:4:0)); for b:=1 to bins do begin r:=round(b*rm/bins);d:=r+r+1; if d Resent-Message-ID: <"n_NVa2.0.u25.mMg2u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1842 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 2087 Status: O >Date: Fri, 15 Oct 1999 12:28:48 -0500 >To: nih-image@io.ece.drexel.edu >From: Kimberly A Howell >Subject: Power Spectrum > >Hello Fellow NIH - Imagers, > > I just have a question for you concerning the power spectrum. > > I want to integrate over all theta at a particular r to obtain the >power spectrum as a function of just r. Does any one know of a >particular macro that would allow me to do that? I have been >investigating the annular selection macro as a way to choose an annular >ROI which would be ideal for what I need to do but that seems to be >useful only for filtering/masking. > > If you could reply directly to me at kahowell@students.wisc.edu that >would be great. Thank you for your time. > >Sincerely, > Kim Howell > >Kimberly Howell, M.S. >University of Wisconsin - Madison >Department of Medical Physics >Phone: (608) 263-9257 > I expect the following does something like what you want? macro '[R]adial Mean'; var w,h,xc,yc,r,rm,d,pi2,c,b,bins:integer; { Greg Joss 1999-10-18} begin GetPicSize(w,h); pi2:=arctan(1)*8; setUser1Label('dArea'); setUser2Label('dIntens'); setOptions('Area Mean User1 User2'); setPrecision(0,5); xc:=w/2; yc:=h/2; rm:=xc; bins:=getNumber('# bins',rm); resetCounter;setCounter(1); if get('maxMeasurements')',d:4:0)); for b:=1 to bins do begin r:=round(b*rm/bins);d:=r+r+1; if d In-Reply-To: Mime-Version: 1.0 Date: Mon, 18 Oct 1999 09:53:24 -0400 To: nih-image@io.ece.drexel.edu From: Wayne Rasband Subject: Re: Grey Level and Profile Plot Resent-Message-ID: <"1hTLq3.0.-W6.qFo2u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1843 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 360 Status: O >May I know why there is a decimal place in the values of the grey level when >a profile is plotted? I thought the grey scale consists of integers from >0-255. Thanks. >Aaron Goh >Imperial College, London The profile plotting routine in NIH Image uses bilinear interpolation. Each plot value is the weighted average of four integer pixel values. -wayne From nih-image-request@io.ece.drexel.edu Mon Oct 18 16:26 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA08836; Mon, 18 Oct 1999 16:26:38 -0400 (EDT) Resent-Date: Mon, 18 Oct 1999 16:26:38 -0400 (EDT) Message-Id: <3.0.5.32.19991018140611.009bcb50@NMT.EDU> X-Sender: grawling@NMT.EDU X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Mon, 18 Oct 1999 14:06:11 -0600 To: nih-image@io.ece.drexel.edu From: "Geoffrey C. Rawling" Subject: grain counting Mime-Version: 1.0 Resent-Message-ID: <"Syqb11.0.1H1.Rxt2u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1844 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1207 Status: O Hi imagers, I have images of sand grains on laying a glass slide. I want to measure particle sizes but one needs to correct for grains intersecting the edges of the image. It seems there are two methods, one in which a guard frame is used in the image to exclude a portion of the image and one in which grain counts are corrected based on their size using their caliper dimensions (image processing handbook v3, pg529). My question is, can one use the dimensions of the fitted ellipse calculated in "analyze particles" as an approximation to the caliper dimensions of the particle if you project the ellipse dimensions onto the x and y axes? And how exactly is the best fit ellipse calculated? My particles are all convex and fairly rounded, so it seems the error may not be too great. Alternatively, I could make a guard frame region and only select those grains within it using "feature-and", but the routine I have for the this operation is pretty slow. Any thoughts are appreciated! Thanks, Geff ************************************* Geoffrey C. Rawling Earth and Environmental Science Dept. New Mexico Tech Socorro, NM 87801 (505)-835-5952 grawling@nmt.edu ************************************* From nih-image-request@io.ece.drexel.edu Mon Oct 18 19:41 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA13650; Mon, 18 Oct 1999 19:41:59 -0400 (EDT) Resent-Date: Mon, 18 Oct 1999 19:41:59 -0400 (EDT) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: grawling@nmt.edu, nih-image@io.ece.drexel.edu Date: Tue, 19 Oct 1999 9:36:03 +1000 Subject: Re: grain counting Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <1F5B4734FAA@rna.bio.mq.edu.au> Resent-Message-ID: <"jDxiy2.0.lp2.fsw2u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1845 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 3940 Status: O >Date: Mon, 18 Oct 1999 14:06:11 -0600 >To: nih-image@io.ece.drexel.edu >From: "Geoffrey C. Rawling" >Subject: grain counting > > Hi imagers, > > I have images of sand grains on laying a glass slide. I want to >measure particle sizes but one needs to correct for grains intersecting >the edges of the image. It seems there are two methods, one in which a >guard frame is used in the image to exclude a portion of the image and >one in which grain counts are corrected based on their size using their >caliper dimensions (image processing handbook v3, pg529). My question >is, can one use the dimensions of the fitted ellipse calculated in >"analyze particles" as an approximation to the caliper dimensions of >the particle if you project the ellipse dimensions onto the x and y >axes? And how exactly is the best fit ellipse calculated? My particles >are all convex and fairly rounded, so it seems the error may not be too >great. Alternatively, I could make a guard frame region and only select >those grains within it using "feature-and", but the routine I have for >the this operation is pretty slow. Any thoughts are appreciated! > >Thanks, Geff >************************************* >Geoffrey C. Rawling >Earth and Environmental Science Dept. >New Mexico Tech >Socorro, NM 87801 >(505)-835-5952 >grawling@nmt.edu >************************************* Geff, There was a thread just last week that discussed methods to drop particles touching the edges of the image. The macro below avoids the problems raised in that thread. Ellipse major and minor axis are useful to detect/elliminate outlier "particles" such as contacting pairs from otherwise relatively uniform clusters. My recollection is that the 'best fit ellipse' algorithm is well documented in the source code... a major advantage in public domain software over proprietry packages. Rather than fit to the outline of the particle, the 'best fit ellipse' is designed to match the moment of inertia of the binary particle with an equivalent elipse; so for your particles works well. For odd particles such as dumbell shapes, the boundaries would not match well. Perimeter/area ratios also help to discriminate between particles. Using the 'autoOutline' as below, you can locate and do detail analysis of the boundaries of each particle. If you wish to make much use of autoOutline, I suggest you use the NIH-Image extention: Object-Image by Norbert Vischer, http://simon.bio.uva.nl/object-image.html which returns coordinates for particles designed to operate with autoOutline. ___________________________ procedure dropEdgeTouchers; { Greg Joss 1999-10-14 assumes binary image input; labels pixels touching boundary with 128; finds label particles (as analyzed Particles were all vertical or horizontal edge strokes then autoOutline will find each particle without fail); converts whole particles connected to labels to all label 128; (needed to leave particles that touched edge because may touch more than once autoOutline must find each time); removes all labeled particles } var i,pid:integer; begin pid:=pidNumber; duplicate('temp1');duplicate('temp2'); setBackground(0);setForeground(128); selectAll;drawBoundary;copy;dispose; selectPic(nPics);paste;doAnd; selectAll;copy;dispose; selectPic(pid);paste;doReplace; setOptions('X-Y Center'); setDensitySlice(128,128); setParticleSize(1,999999); analyzeParticles('reset');undo; setDensitySlice(0,0); setThreshold(128); setForeground(128); for i:=1 to rCount do begin autoOutline(rX[i],rY[i]); fill; end; setThreshold(-1); selectAll; changeValues(128,128,0); setForeground(255); end macro'/0 dropEdgeTouchers';begin dropEdgeTouchers; end Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-d-request@io.ece.drexel.edu Tue Oct 19 04:15 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id EAA15095; Tue, 19 Oct 1999 04:15:34 -0400 (EDT) Date: Tue, 19 Oct 1999 04:15:34 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910190815.EAA15095@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #239 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/239 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 8517 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 239 Today's Topics: Re: Grey Level and Profile Plot [ Wayne Rasband ] grain counting [ "Geoffrey C. Rawling" ] ------------------------------ Date: Mon, 18 Oct 1999 09:53:24 -0400 From: Wayne Rasband To: nih-image@io.ece.drexel.edu Subject: Re: Grey Level and Profile Plot Message-Id: Content-Type: text/plain; charset="us-ascii" >May I know why there is a decimal place in the values of the grey level when >a profile is plotted? I thought the grey scale consists of integers from >0-255. Thanks. >Aaron Goh >Imperial College, London The profile plotting routine in NIH Image uses bilinear interpolation. Each plot value is the weighted average of four integer pixel values. -wayne ------------------------------ Date: Mon, 18 Oct 1999 14:06:11 -0600 From: "Geoffrey C. Rawling" To: nih-image@io.ece.drexel.edu Subject: grain counting Message-Id: <3.0.5.32.19991018140611.009bcb50@NMT.EDU> Content-Type: text/plain; charset="us-ascii" Hi imagers, I have images of sand grains on laying a glass slide. I want to measure particle sizes but one needs to correct for grains intersecting the edges of the image. It seems there are two methods, one in which a guard frame is used in the image to exclude a portion of the image and one in which grain counts are corrected based on their size using their caliper dimensions (image processing handbook v3, pg529). My question is, can one use the dimensions of the fitted ellipse calculated in "analyze particles" as an approximation to the caliper dimensions of the particle if you project the ellipse dimensions onto the x and y axes? And how exactly is the best fit ellipse calculated? My particles are all convex and fairly rounded, so it seems the error may not be too great. Alternatively, I could make a guard frame region and only select those grains within it using "feature-and", but the routine I have for the this operation is pretty slow. Any thoughts are appreciated! Thanks, Geff ************************************* Geoffrey C. Rawling Earth and Environmental Science Dept. New Mexico Tech Socorro, NM 87801 (505)-835-5952 grawling@nmt.edu ************************************* ------------------------------ Date: Tue, 19 Oct 1999 9:36:03 +1000 From: GJOSS@rna.bio.mq.edu.au To: grawling@nmt.edu, nih-image@io.ece.drexel.edu Subject: Re: grain counting Message-ID: <1F5B4734FAA@rna.bio.mq.edu.au> >Date: Mon, 18 Oct 1999 14:06:11 -0600 >To: nih-image@io.ece.drexel.edu >From: "Geoffrey C. Rawling" >Subject: grain counting > > Hi imagers, > > I have images of sand grains on laying a glass slide. I want to >measure particle sizes but one needs to correct for grains intersecting >the edges of the image. It seems there are two methods, one in which a >guard frame is used in the image to exclude a portion of the image and >one in which grain counts are corrected based on their size using their >caliper dimensions (image processing handbook v3, pg529). My question >is, can one use the dimensions of the fitted ellipse calculated in >"analyze particles" as an approximation to the caliper dimensions of >the particle if you project the ellipse dimensions onto the x and y >axes? And how exactly is the best fit ellipse calculated? My particles >are all convex and fairly rounded, so it seems the error may not be too >great. Alternatively, I could make a guard frame region and only select >those grains within it using "feature-and", but the routine I have for >the this operation is pretty slow. Any thoughts are appreciated! > >Thanks, Geff >************************************* >Geoffrey C. Rawling >Earth and Environmental Science Dept. >New Mexico Tech >Socorro, NM 87801 >(505)-835-5952 >grawling@nmt.edu >************************************* Geff, There was a thread just last week that discussed methods to drop particles touching the edges of the image. The macro below avoids the problems raised in that thread. Ellipse major and minor axis are useful to detect/elliminate outlier "particles" such as contacting pairs from otherwise relatively uniform clusters. My recollection is that the 'best fit ellipse' algorithm is well documented in the source code... a major advantage in public domain software over proprietry packages. Rather than fit to the outline of the particle, the 'best fit ellipse' is designed to match the moment of inertia of the binary particle with an equivalent elipse; so for your particles works well. For odd particles such as dumbell shapes, the boundaries would not match well. Perimeter/area ratios also help to discriminate between particles. Using the 'autoOutline' as below, you can locate and do detail analysis of the boundaries of each particle. If you wish to make much use of autoOutline, I suggest you use the NIH-Image extention: Object-Image by Norbert Vischer, http://simon.bio.uva.nl/object-image.html which returns coordinates for particles designed to operate with autoOutline. ___________________________ procedure dropEdgeTouchers; { Greg Joss 1999-10-14 assumes binary image input; labels pixels touching boundary with 128; finds label particles (as analyzed Particles were all vertical or horizontal edge strokes then autoOutline will find each particle without fail); converts whole particles connected to labels to all label 128; (needed to leave particles that touched edge because may touch more than once autoOutline must find each time); removes all labeled particles } var i,pid:integer; begin pid:=pidNumber; duplicate('temp1');duplicate('temp2'); setBackground(0);setForeground(128); selectAll;drawBoundary;copy;dispose; selectPic(nPics);paste;doAnd; selectAll;copy;dispose; selectPic(pid);paste;doReplace; setOptions('X-Y Center'); setDensitySlice(128,128); setParticleSize(1,999999); analyzeParticles('reset');undo; setDensitySlice(0,0); setThreshold(128); setForeground(128); for i:=1 to rCount do begin autoOutline(rX[i],rY[i]); fill; end; setThreshold(-1); selectAll; changeValues(128,128,0); setForeground(255); end macro'/0 dropEdgeTouchers';begin dropEdgeTouchers; end Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Tue, 19 Oct 1999 09:57:30 +0200 From: Ard Jonker To: nih-image@io.ece.drexel.edu Subject: Announcement: DV capture program available! Message-Id: Content-Type: text/plain; charset="us-ascii" Dear List members, I came across a program that does the long-awaited capturing of digital video. Output of the program is a Quicktime movie when recording movies, or PICT files when recording stills (stills can be taken from moving footage). The program is shareware and costs only $15 US. The program , BTV, Written by Ben Bird, benbird@softhome.net can be downloaded from the BTV Home page: http://www.btv.org.uk/ One small remark: connect a camera and feed images to the DV connector (of, eg. your B/W G3) before starting the program, or you will get an error message indicating that hardware could not be initialised. The program seems to be sluggish in displaying your footage, but that's a wrong impression: once you record your film, the data will all be there as far as I could see. Even the audio tracks are perfectly there, once you have set 'audio preferences' to use DV audio as input. I really love the program, and I think it is worth every cent of the shareware fee. I have no connection with mr. Bird. Regards, Ard Jonker dr A.Jonker |Academic Medical Centre Dep of Cell Biology and Histology |Meibergdreef 15 Faculty of Medicine |1105 AZ Amsterdam University of Amsterdam |The Netherlands tel +31 20 566 6804 |fax +31 20 697 4156 PGP fingerprint B263 BEAF 48E2 FE6E 2525 B8C3 EBF9 7685 9A1E 8A3A -------------------------------- End of nih-image-d Digest V99 Issue #239 **************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 19 04:16 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id EAA15291; Tue, 19 Oct 1999 04:16:25 -0400 (EDT) Resent-Date: Tue, 19 Oct 1999 04:16:25 -0400 (EDT) X-Sender: ajonker.public.amc@reddwarf.amc.uva.nl Message-Id: In-Reply-To: <199910180446.AAA21171@io.ece.drexel.edu> Mime-Version: 1.0 Date: Tue, 19 Oct 1999 09:57:30 +0200 To: nih-image@io.ece.drexel.edu From: Ard Jonker Subject: Announcement: DV capture program available! Resent-Message-ID: <"eqMsC2.0.7o2.-H23u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1846 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1387 Status: O Dear List members, I came across a program that does the long-awaited capturing of digital video. Output of the program is a Quicktime movie when recording movies, or PICT files when recording stills (stills can be taken from moving footage). The program is shareware and costs only $15 US. The program , BTV, Written by Ben Bird, benbird@softhome.net can be downloaded from the BTV Home page: http://www.btv.org.uk/ One small remark: connect a camera and feed images to the DV connector (of, eg. your B/W G3) before starting the program, or you will get an error message indicating that hardware could not be initialised. The program seems to be sluggish in displaying your footage, but that's a wrong impression: once you record your film, the data will all be there as far as I could see. Even the audio tracks are perfectly there, once you have set 'audio preferences' to use DV audio as input. I really love the program, and I think it is worth every cent of the shareware fee. I have no connection with mr. Bird. Regards, Ard Jonker dr A.Jonker |Academic Medical Centre Dep of Cell Biology and Histology |Meibergdreef 15 Faculty of Medicine |1105 AZ Amsterdam University of Amsterdam |The Netherlands tel +31 20 566 6804 |fax +31 20 697 4156 PGP fingerprint B263 BEAF 48E2 FE6E 2525 B8C3 EBF9 7685 9A1E 8A3A From nih-image-request@io.ece.drexel.edu Tue Oct 19 07:16 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA17865; Tue, 19 Oct 1999 07:16:01 -0400 (EDT) Resent-Date: Tue, 19 Oct 1999 07:16:01 -0400 (EDT) Message-Id: <199910191059.LAA03100@holyrood.ed.ac.uk> X-Mailer: Microsoft Outlook Express Macintosh Edition - 4.5 (0410) Date: Tue, 19 Oct 1999 11:59:01 +0100 Subject: Re: grain counting From: "Jeremy Brown" To: nih-image@io.ece.drexel.edu Mime-version: 1.0 X-Priority: 3 Content-transfer-encoding: 7bit Resent-Message-ID: <"SXYMA3.0.Ha3.xx43u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1847 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 2599 Status: O > Hi imagers, > > I have images of sand grains on laying a glass slide. I want to measure > particle sizes but one needs to correct for grains intersecting the edges > of the image. It seems there are two methods, one in which a guard frame is > used in the image to exclude a portion of the image and one in which grain > counts are corrected based on their size using their caliper dimensions > (image processing handbook v3, pg529). My question is, can one use the > dimensions of the fitted ellipse calculated in "analyze particles" as an > approximation to the caliper dimensions of the particle if you project the > ellipse dimensions onto the x and y axes? And how exactly is the best fit > ellipse calculated? My particles are all convex and fairly rounded, so it > seems the error may not be too great. Alternatively, I could make a guard > frame region and only select those grains within it using "feature-and", > but the routine I have for the this operation is pretty slow. Any thoughts > are appreciated! > > Thanks, Geff Try Inclusion Counter v1.1, even if I suggest so myself ;-) Handling of colour segmentation is fairly basic in this version but it has a counting frame. In this case the frame is used three times to: 1) calculate % area; 2) calculate objects per field by eliminating objects touching the right hand half of of ROI; and 3) calculate mean object area and STDEV from the mean by eliminating all the objects which touch the ROI boundary. The idea behind it is to automate this kind of process (originally for Chlamydia/Chlamydiophila inclusion bodies), so Images need to be numbered sequentially in the format: image.0001......image.nnnn. It should work under Scion Image with some alterations to the file path nomenclature used in "ApplyROI;" and "Set Region of Interest". You will need a decompression utility which recognises Stuffit archives to download it. Application of the guard frame/measurement of PAL images takes <4 seconds per image of a 350 MHz G3 to about 25 seconds per image on an 80MHz 601 PPC. Which is adequate for most purposes, especially since your not required to make any adjustments whilst it does this. It can be downloaded from http://users.netmatters.co.uk/brownj/ICHomePage.html Regards, Jeremy **************************************** Jeremy Brown Research Fellow Department of Veterinary Clinical Studies Royal (Dick) School of Veterinary Studies Easter Bush Veterinary Centre Easter Bush Roslin Midlothian EH25 9RG Scotland UK http://users.netmatters.co.uk/brownj/HomePage.html **************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 19 11:09 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA00821; Tue, 19 Oct 1999 11:09:07 -0400 (EDT) Resent-Date: Tue, 19 Oct 1999 11:09:07 -0400 (EDT) Subject: Re: grain counting Date: Tue, 19 Oct 99 10:47:51 -0400 x-sender: jlr$biol@qc1.qc.edu x-mailer: Claris Emailer 1.1 From: jared rifkin To: "nih-image users group" Mime-Version: 1.0 Message-ID: <1A6BAE87DC9@qc1.qc.edu> Resent-Message-ID: <"N_P842.0.kj6.rI83u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1848 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 1618 Status: O it seems to me that there are two problems associated with your particle size problem. one- (1) perhaps an implied problem but-how many particles are there? and (2) what is the average size (area? or volume?) of the population? the size problem does not need any consideration of particles not totally visible and there are many geometric analyses presented in the cell biology literature of the '20's through the 40's that deal with cells, read 'particles', of various shapes these should be sufficient . sand grain counting in a bounded field seems entirely analogous to cell counting for biologists. the standard (used by hematologists and others interested in such population measurements) is to assume that (1) all the particles are of the same or reasonably close to the same size, (2) that they are randomly distributed in the field of observation, (3) just as many particles are touching any one boundary as are touching any other single boundary, and (4) that the proportions of particle-touching any boundary line are equally randomized in space. then-- count any particle touching (to any extent) the left and the top boundary lines of the observed field and ignore any particle touching (to any extent) the right and bottom boundary lines. the key preparatory concern is to get a spatially random spread of the particles and count at least 6 (for statistically significant data) different non-touching equal-sized fields. cell counters use a device called a hemacytometer for preparation. given the 'macro' size of sand grains, a scaled-up version will have to be constructed. From nih-image-request@io.ece.drexel.edu Tue Oct 19 12:08 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA10745; Tue, 19 Oct 1999 12:08:17 -0400 (EDT) Resent-Date: Tue, 19 Oct 1999 12:08:17 -0400 (EDT) From: DrJohnRuss@aol.com Message-ID: <0.5003e387.253ded4a@aol.com> Date: Tue, 19 Oct 1999 11:50:34 EDT Subject: Re: grain counting To: nih-image@io.ece.drexel.edu MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Mailer: AOL for Macintosh sub 56 Resent-Message-ID: <"nFgUa.0.5s1.kD93u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1849 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1586 Status: O In a message dated 10/19/99 11:01:21 AM, jared_rifkin@qc.edu writes: >sand grain counting in a bounded field seems entirely analogous to cell > >counting for biologists. >the standard (used by hematologists and others interested in such >population measurements) is to assume that (1) all the particles are of >the same or reasonably close to the same size, (2) that they are randomly >distributed in the field of observation, (3) just as many particles are >touching any one boundary as are touching any other single boundary, and >(4) that the proportions of particle-touching any boundary line are >equally randomized in space. Alas, the assumptions here are really very limiting. The case that raised the original question was specifically dealing with violations of your assumption #1. If the particles/cells/features etc. are NOT all the same size, then large ones will be more likely to intersect the edge of the image and your size distirbution will be biased because large features are undercounted. The two methods of edge correction that were discussed in the original post are both able to correct for this. The question in the post was essentially how to implement them in Image, not how to avoid making the correction. My personal preference is to use the "adjusted count" technique since it avoids discarding features in the guard frame, but it requires determine the X- and Y-Feret's diameters of each feature. I haven't done this specifically with Image but the minimum and maximum X and Y coordinates should be easy enough to extract. John Russ From nih-image-request@io.ece.drexel.edu Tue Oct 19 12:23 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA13636; Tue, 19 Oct 1999 12:23:17 -0400 (EDT) Resent-Date: Tue, 19 Oct 1999 12:23:17 -0400 (EDT) Message-ID: <380C9668.FF608820@lemming0.lem.uni-karlsruhe.de> Date: Tue, 19 Oct 1999 18:03:52 +0200 From: reznik X-Mailer: Mozilla 4.6 [de] (Win95; I) X-Accept-Language: de MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: [Fwd: unsubscribe] Resent-Message-ID: <"fXcLq.0.2M2.1P93u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1850 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: multipart/mixed; boundary="------------3ECE19A5406F51976C0242F2" Content-Length: 1776 Status: O Dies ist eine mehrteilige Nachricht im MIME-Format. --------------3ECE19A5406F51976C0242F2 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 7bit I want to unsubscibe Thanks,B.Reznik --------------3ECE19A5406F51976C0242F2 Content-Type: message/rfc822 Content-Transfer-Encoding: 7bit Content-Disposition: inline Return-Path: Received: from mailgate.rz.uni-karlsruhe.de (really [129.13.197.4]) by lemming0.lem.uni-karlsruhe.de via in.smtpd with esmtp id (Debian Smail3.2.0.102) for ; Sat, 18 Sep 1999 11:13:22 +0100 (BST) Received: from io.ece.drexel.edu (lists@io.ece.drexel.edu [144.118.32.3]) by mailgate.rz.uni-karlsruhe.de with esmtp (Exim 3.02 #2) id 11SHUQ-0000Uu-00; Sat, 18 Sep 1999 12:13:06 +0200 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA13458; Sat, 18 Sep 1999 06:12:48 -0400 (EDT) Resent-Date: Sat, 18 Sep 1999 06:12:48 -0400 (EDT) Message-Id: <3.0.1.32.19990918121259.00733ed0@mail.zedat.fu-berlin.de> X-Sender: mbrehm@mail.zedat.fu-berlin.de X-Mailer: Windows Eudora Pro Version 3.0.1 (32) [G] Date: Sat, 18 Sep 1999 12:12:59 +0200 To: nih-image@io.ece.drexel.edu From: Brehm Subject: unsubscribe Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"k3K7j1.0.5I3.VMsut"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1734 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Resent-Bcc: I want to unsubscribe Thanks M Brehm --------------3ECE19A5406F51976C0242F2-- From nih-image-request@io.ece.drexel.edu Tue Oct 19 12:56 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA19217; Tue, 19 Oct 1999 12:56:02 -0400 (EDT) Resent-Date: Tue, 19 Oct 1999 12:56:02 -0400 (EDT) X-Sender: sbudick1@swarthmore.edu (Unverified) Message-Id: In-Reply-To: <1F5B4734FAA@rna.bio.mq.edu.au> Mime-Version: 1.0 Date: Tue, 19 Oct 1999 12:35:58 -0400 To: nih-image@io.ece.drexel.edu From: Seth Budick Subject: Re: grain counting Resent-Message-ID: <"K3VQW2.0.s04.tv93u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1851 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 883 Status: O Dear NIH-ers, Please forgive my naivete regarding image. I am working with stacks of oval ROI's that have been copy and pasted into the rectangular windows and thus I cannot use ignore particles touching edge since the ROI does not actually contact the window's edge. Becasuse my images are of circular wells, there are often slight shadows along the edges that are inevitably counted as particles when I increase the density slicing sufficiently to capture all of my particles. The idea of excluding particles based on their perimeter/area seems like it might work, but how can you tell Image to exclude particles based on a criterion like that? Thanks in advance. Seth ________________________________ Seth Budick Biology Department 414 Mugar Hall Northeastern University Boston, MA 02115 Phone: (617) 373-4034 e-mail: sbudick1@swarthmore.edu http://www.omalleylab.neu.edu From nih-image-request@io.ece.drexel.edu Tue Oct 19 20:33 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id UAA28093; Tue, 19 Oct 1999 20:33:06 -0400 (EDT) Resent-Date: Tue, 19 Oct 1999 20:33:06 -0400 (EDT) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: sbudick1@swarthmore.edu, nih-image@io.ece.drexel.edu Date: Wed, 20 Oct 1999 10:26:27 +1000 Subject: Re: grain counting Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <20E914D50D8@rna.bio.mq.edu.au> Resent-Message-ID: <"yRQ1I1.0.vQ6.diG3u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1852 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 2612 Status: O >Date: Tue, 19 Oct 1999 12:35:58 -0400 >To: nih-image@io.ece.drexel.edu >From: Seth Budick >Subject: Re: grain counting > >Dear NIH-ers, > Please forgive my naivete regarding image. I am working >with stacks of oval ROI's that have been copy and pasted into the >rectangular windows and thus I cannot use ignore particles touching >edge since the ROI does not actually contact the window's edge. >Becasuse my images are of circular wells, there are often slight >shadows along the edges that are inevitably counted as particles when I >increase the density slicing sufficiently to capture all of my >particles. The idea of excluding particles based on their >perimeter/area seems like it might work, but how can you tell Image to >exclude particles based on a criterion like that? Thanks in advance. > >Seth > >________________________________ >Seth Budick >Biology Department >414 Mugar Hall >Northeastern University >Boston, MA 02115 > >Phone: (617) 373-4034 >e-mail: sbudick1@swarthmore.edu >http://www.omalleylab.neu.edu Seth, The macro I posted yesterday can be simply modified to cater for touching a boundary of any shape. The 'dropEdgeTouchers' procedure used selectAll;drawBoundary; to create a one pixel wide boundary mask. By getPicSize(w,h);setThreshold(1);autoOutline(0,h/2);drawBoundary; you can recover the ROI used to make the original paste. Then the same procedure would work to remove particles touching the boundary except for some possible problems with autoOutline(rX[i],rY[i]); as the centroid of the particle might possibly not fall within the particle. You can code to detect and correct for this in NIH-Image or simply use Object-Image and rTop[],rLeft[] results instead of rX[],rY[]. Excluding particles based on their perimeter/area generally works quite well. The trick is that you cant exclude on the criteron but allow Image to measure all particles in the ROI and then do post analysis of the results table to mark qualifying/failing particles by code such as: {note that for particleAnalysis rLength[i] is perimeter of particle} for i:=1 to rCount do if rArea[i]/rLength[i]>ratioLimit then rUser1[i]:=0 else rUser1[i]:=1; If you need to, you can effectively (if crudely) delete non qualifying particles from the results table (rather than simple marking) by shuffling row contents down the table and setCounter() as a final pass. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-d-request@io.ece.drexel.edu Tue Oct 19 23:56 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id XAA25589; Tue, 19 Oct 1999 23:56:18 -0400 (EDT) Date: Tue, 19 Oct 1999 23:56:18 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910200356.XAA25589@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #240 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/240 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 14986 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 240 Today's Topics: Re: grain counting [ "Jeremy Brown" ] Re: grain counting [ DrJohnRuss@aol.com ] [Fwd: unsubscribe] [ reznik To: nih-image@io.ece.drexel.edu Subject: Re: grain counting Message-Id: <199910191059.LAA03100@holyrood.ed.ac.uk> Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit > Hi imagers, > > I have images of sand grains on laying a glass slide. I want to measure > particle sizes but one needs to correct for grains intersecting the edges > of the image. It seems there are two methods, one in which a guard frame is > used in the image to exclude a portion of the image and one in which grain > counts are corrected based on their size using their caliper dimensions > (image processing handbook v3, pg529). My question is, can one use the > dimensions of the fitted ellipse calculated in "analyze particles" as an > approximation to the caliper dimensions of the particle if you project the > ellipse dimensions onto the x and y axes? And how exactly is the best fit > ellipse calculated? My particles are all convex and fairly rounded, so it > seems the error may not be too great. Alternatively, I could make a guard > frame region and only select those grains within it using "feature-and", > but the routine I have for the this operation is pretty slow. Any thoughts > are appreciated! > > Thanks, Geff Try Inclusion Counter v1.1, even if I suggest so myself ;-) Handling of colour segmentation is fairly basic in this version but it has a counting frame. In this case the frame is used three times to: 1) calculate % area; 2) calculate objects per field by eliminating objects touching the right hand half of of ROI; and 3) calculate mean object area and STDEV from the mean by eliminating all the objects which touch the ROI boundary. The idea behind it is to automate this kind of process (originally for Chlamydia/Chlamydiophila inclusion bodies), so Images need to be numbered sequentially in the format: image.0001......image.nnnn. It should work under Scion Image with some alterations to the file path nomenclature used in "ApplyROI;" and "Set Region of Interest". You will need a decompression utility which recognises Stuffit archives to download it. Application of the guard frame/measurement of PAL images takes <4 seconds per image of a 350 MHz G3 to about 25 seconds per image on an 80MHz 601 PPC. Which is adequate for most purposes, especially since your not required to make any adjustments whilst it does this. It can be downloaded from http://users.netmatters.co.uk/brownj/ICHomePage.html Regards, Jeremy **************************************** Jeremy Brown Research Fellow Department of Veterinary Clinical Studies Royal (Dick) School of Veterinary Studies Easter Bush Veterinary Centre Easter Bush Roslin Midlothian EH25 9RG Scotland UK http://users.netmatters.co.uk/brownj/HomePage.html **************************************** ------------------------------ Date: Tue, 19 Oct 99 10:47:51 -0400 From: jared rifkin To: "nih-image users group" Subject: Re: grain counting Message-ID: <1A6BAE87DC9@qc1.qc.edu> Content-Type: text/plain; charset="US-ASCII" it seems to me that there are two problems associated with your particle size problem. one- (1) perhaps an implied problem but-how many particles are there? and (2) what is the average size (area? or volume?) of the population? the size problem does not need any consideration of particles not totally visible and there are many geometric analyses presented in the cell biology literature of the '20's through the 40's that deal with cells, read 'particles', of various shapes these should be sufficient . sand grain counting in a bounded field seems entirely analogous to cell counting for biologists. the standard (used by hematologists and others interested in such population measurements) is to assume that (1) all the particles are of the same or reasonably close to the same size, (2) that they are randomly distributed in the field of observation, (3) just as many particles are touching any one boundary as are touching any other single boundary, and (4) that the proportions of particle-touching any boundary line are equally randomized in space. then-- count any particle touching (to any extent) the left and the top boundary lines of the observed field and ignore any particle touching (to any extent) the right and bottom boundary lines. the key preparatory concern is to get a spatially random spread of the particles and count at least 6 (for statistically significant data) different non-touching equal-sized fields. cell counters use a device called a hemacytometer for preparation. given the 'macro' size of sand grains, a scaled-up version will have to be constructed. ------------------------------ Date: Tue, 19 Oct 1999 11:50:34 EDT From: DrJohnRuss@aol.com To: nih-image@io.ece.drexel.edu Subject: Re: grain counting Message-ID: <0.5003e387.253ded4a@aol.com> Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit In a message dated 10/19/99 11:01:21 AM, jared_rifkin@qc.edu writes: >sand grain counting in a bounded field seems entirely analogous to cell > >counting for biologists. >the standard (used by hematologists and others interested in such >population measurements) is to assume that (1) all the particles are of >the same or reasonably close to the same size, (2) that they are randomly >distributed in the field of observation, (3) just as many particles are >touching any one boundary as are touching any other single boundary, and >(4) that the proportions of particle-touching any boundary line are >equally randomized in space. Alas, the assumptions here are really very limiting. The case that raised the original question was specifically dealing with violations of your assumption #1. If the particles/cells/features etc. are NOT all the same size, then large ones will be more likely to intersect the edge of the image and your size distirbution will be biased because large features are undercounted. The two methods of edge correction that were discussed in the original post are both able to correct for this. The question in the post was essentially how to implement them in Image, not how to avoid making the correction. My personal preference is to use the "adjusted count" technique since it avoids discarding features in the guard frame, but it requires determine the X- and Y-Feret's diameters of each feature. I haven't done this specifically with Image but the minimum and maximum X and Y coordinates should be easy enough to extract. John Russ ------------------------------ Date: Tue, 19 Oct 1999 18:03:52 +0200 From: reznik To: nih-image@io.ece.drexel.edu Subject: [Fwd: unsubscribe] Message-ID: <380C9668.FF608820@lemming0.lem.uni-karlsruhe.de> Content-Type: multipart/mixed; boundary="------------3ECE19A5406F51976C0242F2" Dies ist eine mehrteilige Nachricht im MIME-Format. --------------3ECE19A5406F51976C0242F2 Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 7bit I want to unsubscibe Thanks,B.Reznik --------------3ECE19A5406F51976C0242F2 Content-Type: message/rfc822 Content-Transfer-Encoding: 7bit Content-Disposition: inline Return-Path: Received: from mailgate.rz.uni-karlsruhe.de (really [129.13.197.4]) by lemming0.lem.uni-karlsruhe.de via in.smtpd with esmtp id (Debian Smail3.2.0.102) for ; Sat, 18 Sep 1999 11:13:22 +0100 (BST) Received: from io.ece.drexel.edu (lists@io.ece.drexel.edu [144.118.32.3]) by mailgate.rz.uni-karlsruhe.de with esmtp (Exim 3.02 #2) id 11SHUQ-0000Uu-00; Sat, 18 Sep 1999 12:13:06 +0200 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA13458; Sat, 18 Sep 1999 06:12:48 -0400 (EDT) Resent-Date: Sat, 18 Sep 1999 06:12:48 -0400 (EDT) Message-Id: <3.0.1.32.19990918121259.00733ed0@mail.zedat.fu-berlin.de> X-Sender: mbrehm@mail.zedat.fu-berlin.de X-Mailer: Windows Eudora Pro Version 3.0.1 (32) [G] Date: Sat, 18 Sep 1999 12:12:59 +0200 To: nih-image@io.ece.drexel.edu From: Brehm Subject: unsubscribe Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Resent-Message-ID: <"k3K7j1.0.5I3.VMsut"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1734 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Resent-Bcc: I want to unsubscribe Thanks M Brehm --------------3ECE19A5406F51976C0242F2-- ------------------------------ Date: Tue, 19 Oct 1999 12:35:58 -0400 From: Seth Budick To: nih-image@io.ece.drexel.edu Subject: Re: grain counting Message-Id: Content-Type: text/plain; charset="us-ascii" Dear NIH-ers, Please forgive my naivete regarding image. I am working with stacks of oval ROI's that have been copy and pasted into the rectangular windows and thus I cannot use ignore particles touching edge since the ROI does not actually contact the window's edge. Becasuse my images are of circular wells, there are often slight shadows along the edges that are inevitably counted as particles when I increase the density slicing sufficiently to capture all of my particles. The idea of excluding particles based on their perimeter/area seems like it might work, but how can you tell Image to exclude particles based on a criterion like that? Thanks in advance. Seth ________________________________ Seth Budick Biology Department 414 Mugar Hall Northeastern University Boston, MA 02115 Phone: (617) 373-4034 e-mail: sbudick1@swarthmore.edu http://www.omalleylab.neu.edu ------------------------------ Date: Wed, 20 Oct 1999 10:26:27 +1000 From: GJOSS@rna.bio.mq.edu.au To: sbudick1@swarthmore.edu, nih-image@io.ece.drexel.edu Subject: Re: grain counting Message-ID: <20E914D50D8@rna.bio.mq.edu.au> >Date: Tue, 19 Oct 1999 12:35:58 -0400 >To: nih-image@io.ece.drexel.edu >From: Seth Budick >Subject: Re: grain counting > >Dear NIH-ers, > Please forgive my naivete regarding image. I am working >with stacks of oval ROI's that have been copy and pasted into the >rectangular windows and thus I cannot use ignore particles touching >edge since the ROI does not actually contact the window's edge. >Becasuse my images are of circular wells, there are often slight >shadows along the edges that are inevitably counted as particles when I >increase the density slicing sufficiently to capture all of my >particles. The idea of excluding particles based on their >perimeter/area seems like it might work, but how can you tell Image to >exclude particles based on a criterion like that? Thanks in advance. > >Seth > >________________________________ >Seth Budick >Biology Department >414 Mugar Hall >Northeastern University >Boston, MA 02115 > >Phone: (617) 373-4034 >e-mail: sbudick1@swarthmore.edu >http://www.omalleylab.neu.edu Seth, The macro I posted yesterday can be simply modified to cater for touching a boundary of any shape. The 'dropEdgeTouchers' procedure used selectAll;drawBoundary; to create a one pixel wide boundary mask. By getPicSize(w,h);setThreshold(1);autoOutline(0,h/2);drawBoundary; you can recover the ROI used to make the original paste. Then the same procedure would work to remove particles touching the boundary except for some possible problems with autoOutline(rX[i],rY[i]); as the centroid of the particle might possibly not fall within the particle. You can code to detect and correct for this in NIH-Image or simply use Object-Image and rTop[],rLeft[] results instead of rX[],rY[]. Excluding particles based on their perimeter/area generally works quite well. The trick is that you cant exclude on the criteron but allow Image to measure all particles in the ROI and then do post analysis of the results table to mark qualifying/failing particles by code such as: {note that for particleAnalysis rLength[i] is perimeter of particle} for i:=1 to rCount do if rArea[i]/rLength[i]>ratioLimit then rUser1[i]:=0 else rUser1[i]:=1; If you need to, you can effectively (if crudely) delete non qualifying particles from the results table (rather than simple marking) by shuffling row contents down the table and setCounter() as a final pass. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Tue, 19 Oct 1999 13:20:03 GMT From: "Enrico BONINO" To: nih-image@io.ece.drexel.edu Subject: Re: Publishing/publicising macros Message-ID: <19991019132004.88972.qmail@hotmail.com> Content-Type: text/plain; format=flowed Gareth Are the macros availables via the web? At what address? What kind of operations are availables on your macros? I need to make geometric corrections (1, 2 and 3 order), classification, georeferencering (tifw) and other operations on satellital and aerial photo's. In the same way I'm working on DEM's to make 3D representation (with Image SXM & photoshop). I'm interested to see how they work, because the GIS softwares that are installed in our laboratory (ArcView, ERDASImagine) need the hardware key or Unix systems, and for "extra-work-at-home" is clearly impossible. Enrico Thanks ************************************************************ Enrico BONINO, geologist University of Liege Dept. of Applied Geology Lab. MICA Geomaterials Characterization - bat.D2 Avenue des Tilleuls, 45 B-4000 LIEGE BELGIUM tel. 0032 (0)4 3669571 fax 0032 (0)4 3669520 http://www.ulg.ac.be/mica E-mail: e_bonino@hotmail.com CV-> http://ewse.ceo.org/anonymous/construct/build.pl/690804 ************************************************************ ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com -------------------------------- End of nih-image-d Digest V99 Issue #240 **************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 19 23:58 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id XAA25958; Tue, 19 Oct 1999 23:58:01 -0400 (EDT) Resent-Date: Tue, 19 Oct 1999 23:58:01 -0400 (EDT) Message-ID: <19991019132004.88972.qmail@hotmail.com> X-Originating-IP: [193.190.182.11] From: "Enrico BONINO" To: nih-image@io.ece.drexel.edu Subject: Re: Publishing/publicising macros Date: Tue, 19 Oct 1999 13:20:03 GMT Mime-Version: 1.0 Resent-Message-ID: <"7YeS82.0.Gp5.jhJ3u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1853 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; format=flowed Content-Length: 1165 Status: O Gareth Are the macros availables via the web? At what address? What kind of operations are availables on your macros? I need to make geometric corrections (1, 2 and 3 order), classification, georeferencering (tifw) and other operations on satellital and aerial photo's. In the same way I'm working on DEM's to make 3D representation (with Image SXM & photoshop). I'm interested to see how they work, because the GIS softwares that are installed in our laboratory (ArcView, ERDASImagine) need the hardware key or Unix systems, and for "extra-work-at-home" is clearly impossible. Enrico Thanks ************************************************************ Enrico BONINO, geologist University of Liege Dept. of Applied Geology Lab. MICA Geomaterials Characterization - bat.D2 Avenue des Tilleuls, 45 B-4000 LIEGE BELGIUM tel. 0032 (0)4 3669571 fax 0032 (0)4 3669520 http://www.ulg.ac.be/mica E-mail: e_bonino@hotmail.com CV-> http://ewse.ceo.org/anonymous/construct/build.pl/690804 ************************************************************ ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Wed Oct 20 04:34 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id EAA01888; Wed, 20 Oct 1999 04:34:46 -0400 (EDT) Resent-Date: Wed, 20 Oct 1999 04:34:46 -0400 (EDT) Message-Id: <199910200821.JAA08990@holyrood.ed.ac.uk> X-Mailer: Microsoft Outlook Express Macintosh Edition - 4.5 (0410) Date: Wed, 20 Oct 1999 09:21:43 +0100 Subject: Re: grain counting (John Russ) From: "Jeremy Brown" To: nih-image@io.ece.drexel.edu Mime-version: 1.0 X-Priority: 3 Content-transfer-encoding: 7bit Resent-Message-ID: <"nq63Y2.0.EJ7.UkN3u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1854 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 3010 Status: O I have just added adjusted count output to Inclusion Counter based on X- and Y- Feret measurements. Implementation of the approach described by John Russ (IPHB 2nd ed pp.503) is remarkable simple and quick in Image. I've also applied this approach to mean area and STDEV from the mean area on the grounds that if any given object should be counted, for example, 1.5 times then it's area value (for purposes of calculating the mean) should be likewise adjusted before calculation of the mean. STDEV has been treated in the same way. I think this approach is correct but I could be wrong. For the purpose of counting objects of similar size (ie low STDEV from mean area) original "haematocytometer" based measurements have been preserved. I will publish the information on my WEB page in the next couple of hours, feedback on the implementation of the equation for adjusted count/area would be greatly appreciated. Jeremy **************************************** Jeremy Brown Research Fellow Department of Veterinary Clinical Studies Royal (Dick) School of Veterinary Studies Easter Bush Veterinary Centre Easter Bush Roslin Midlothian EH25 9RG Scotland UK http://users.netmatters.co.uk/brownj/HomePage.html **************************************** ---------- >From: DrJohnRuss@aol.com >To: nih-image@io.ece.drexel.edu >Subject: Re: grain counting >Date: Tue, Oct 19, 1999, 4:50 pm > > > In a message dated 10/19/99 11:01:21 AM, jared_rifkin@qc.edu writes: > >>sand grain counting in a bounded field seems entirely analogous to cell >> >>counting for biologists. >>the standard (used by hematologists and others interested in such >>population measurements) is to assume that (1) all the particles are of >>the same or reasonably close to the same size, (2) that they are randomly >>distributed in the field of observation, (3) just as many particles are >>touching any one boundary as are touching any other single boundary, and >>(4) that the proportions of particle-touching any boundary line are >>equally randomized in space. > > Alas, the assumptions here are really very limiting. The case that raised the > original question was specifically dealing with violations of your assumption > #1. If the particles/cells/features etc. are NOT all the same size, then > large ones will be more likely to intersect the edge of the image and your > size distirbution will be biased because large features are undercounted. The > two methods of edge correction that were discussed in the original post are > both able to correct for this. The question in the post was essentially how > to implement them in Image, not how to avoid making the correction. My > personal preference is to use the "adjusted count" technique since it avoids > discarding features in the guard frame, but it requires determine the X- and > Y-Feret's diameters of each feature. I haven't done this specifically with > Image but the minimum and maximum X and Y coordinates should be easy enough > to extract. > > John Russ > From nih-image-request@io.ece.drexel.edu Wed Oct 20 06:22 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA15887; Wed, 20 Oct 1999 06:22:27 -0400 (EDT) Resent-Date: Wed, 20 Oct 1999 06:22:27 -0400 (EDT) Date: Wed, 20 Oct 1999 11:07:29 +0100 (BST) From: Boris Reznik To: nih-image@io.ece.drexel.edu Subject: Re: unsubscribe In-Reply-To: <380C9668.FF608820@lemming0.lem.uni-karlsruhe.de> Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"K2jVf2.0.-M3.eHP3u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1855 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 50 Status: O > I want to unsubscibe > > Thanks,B.Reznik From nih-image-request@io.ece.drexel.edu Wed Oct 20 09:58 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA13822; Wed, 20 Oct 1999 09:58:51 -0400 (EDT) Resent-Date: Wed, 20 Oct 1999 09:58:51 -0400 (EDT) Message-Id: <199910201342.OAA03157@holyrood.ed.ac.uk> X-Mailer: Microsoft Outlook Express Macintosh Edition - 4.5 (0410) Date: Wed, 20 Oct 1999 14:42:40 +0100 Subject: Re: grain counting (John Russ) From: "Jeremy Brown" To: nih-image@io.ece.drexel.edu Mime-version: 1.0 X-Priority: 3 Content-transfer-encoding: 7bit Resent-Message-ID: <"Ik0wm2.0.zr2.LRS3u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1856 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 407 Status: O The macro is finished and can be downloaded from the website below. Jeremy **************************************** Jeremy Brown Research Fellow Department of Veterinary Clinical Studies Royal (Dick) School of Veterinary Studies Easter Bush Veterinary Centre Easter Bush Roslin Midlothian EH25 9RG Scotland UK http://users.netmatters.co.uk/brownj/HomePage.html **************************************** From nih-image-request@io.ece.drexel.edu Wed Oct 20 14:15 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id OAA21965; Wed, 20 Oct 1999 14:15:12 -0400 (EDT) Resent-Date: Wed, 20 Oct 1999 14:15:12 -0400 (EDT) Mime-Version: 1.0 X-Sender: means@wave22i.nrl.navy.mil Message-Id: Date: Wed, 20 Oct 1999 14:00:30 -0400 To: nih-image@io.ece.drexel.edu From: Steve Means Subject: SetOptions('Minor') problem Resent-Message-ID: <"_CVYG.0.fr4.5DW3u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1857 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 621 Status: O When I use the SetOptions with a string containing 'Minor' it produces additional unrequested options (e.g.; User1, User2, etc.). If I use the identical string, excluding 'Minor', it works properly. e.g.: SetOptions('X-Y Center,Major,Angle,Area'); - Works. SetOptions('X-Y Center,Major,Minor,Angle,Area'); - Does not work. I'm using Image v1.62 on a 450MHz G3. Any suggestions on how to solve this problem would be greatly appreciated. Steve Means Naval Research Laboratory 4555 Overlook Avenue SW Code 7121 Washington, DC 20375 Email: means@wave.nrl.navy.mil Voice: 202-404-4653 Fax: 202-404-7813 From nih-image-d-request@io.ece.drexel.edu Thu Oct 21 06:12 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA27300; Thu, 21 Oct 1999 06:12:20 -0400 (EDT) Date: Thu, 21 Oct 1999 06:12:20 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910211012.GAA27300@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #241 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/241 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5882 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 241 Today's Topics: Re: grain counting (John Russ) [ "Jeremy Brown" To: nih-image@io.ece.drexel.edu Subject: Re: grain counting (John Russ) Message-Id: <199910200821.JAA08990@holyrood.ed.ac.uk> Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit I have just added adjusted count output to Inclusion Counter based on X- and Y- Feret measurements. Implementation of the approach described by John Russ (IPHB 2nd ed pp.503) is remarkable simple and quick in Image. I've also applied this approach to mean area and STDEV from the mean area on the grounds that if any given object should be counted, for example, 1.5 times then it's area value (for purposes of calculating the mean) should be likewise adjusted before calculation of the mean. STDEV has been treated in the same way. I think this approach is correct but I could be wrong. For the purpose of counting objects of similar size (ie low STDEV from mean area) original "haematocytometer" based measurements have been preserved. I will publish the information on my WEB page in the next couple of hours, feedback on the implementation of the equation for adjusted count/area would be greatly appreciated. Jeremy **************************************** Jeremy Brown Research Fellow Department of Veterinary Clinical Studies Royal (Dick) School of Veterinary Studies Easter Bush Veterinary Centre Easter Bush Roslin Midlothian EH25 9RG Scotland UK http://users.netmatters.co.uk/brownj/HomePage.html **************************************** ---------- >From: DrJohnRuss@aol.com >To: nih-image@io.ece.drexel.edu >Subject: Re: grain counting >Date: Tue, Oct 19, 1999, 4:50 pm > > > In a message dated 10/19/99 11:01:21 AM, jared_rifkin@qc.edu writes: > >>sand grain counting in a bounded field seems entirely analogous to cell >> >>counting for biologists. >>the standard (used by hematologists and others interested in such >>population measurements) is to assume that (1) all the particles are of >>the same or reasonably close to the same size, (2) that they are randomly >>distributed in the field of observation, (3) just as many particles are >>touching any one boundary as are touching any other single boundary, and >>(4) that the proportions of particle-touching any boundary line are >>equally randomized in space. > > Alas, the assumptions here are really very limiting. The case that raised the > original question was specifically dealing with violations of your assumption > #1. If the particles/cells/features etc. are NOT all the same size, then > large ones will be more likely to intersect the edge of the image and your > size distirbution will be biased because large features are undercounted. The > two methods of edge correction that were discussed in the original post are > both able to correct for this. The question in the post was essentially how > to implement them in Image, not how to avoid making the correction. My > personal preference is to use the "adjusted count" technique since it avoids > discarding features in the guard frame, but it requires determine the X- and > Y-Feret's diameters of each feature. I haven't done this specifically with > Image but the minimum and maximum X and Y coordinates should be easy enough > to extract. > > John Russ > ------------------------------ Date: Wed, 20 Oct 1999 11:07:29 +0100 (BST) From: Boris Reznik To: nih-image@io.ece.drexel.edu Subject: Re: unsubscribe Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII > I want to unsubscibe > > Thanks,B.Reznik ------------------------------ Date: Wed, 20 Oct 1999 14:42:40 +0100 From: "Jeremy Brown" To: nih-image@io.ece.drexel.edu Subject: Re: grain counting (John Russ) Message-Id: <199910201342.OAA03157@holyrood.ed.ac.uk> Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit The macro is finished and can be downloaded from the website below. Jeremy **************************************** Jeremy Brown Research Fellow Department of Veterinary Clinical Studies Royal (Dick) School of Veterinary Studies Easter Bush Veterinary Centre Easter Bush Roslin Midlothian EH25 9RG Scotland UK http://users.netmatters.co.uk/brownj/HomePage.html **************************************** ------------------------------ Date: Wed, 20 Oct 1999 14:00:30 -0400 From: Steve Means To: nih-image@io.ece.drexel.edu Subject: SetOptions('Minor') problem Message-Id: Content-Type: text/plain; charset="us-ascii" When I use the SetOptions with a string containing 'Minor' it produces additional unrequested options (e.g.; User1, User2, etc.). If I use the identical string, excluding 'Minor', it works properly. e.g.: SetOptions('X-Y Center,Major,Angle,Area'); - Works. SetOptions('X-Y Center,Major,Minor,Angle,Area'); - Does not work. I'm using Image v1.62 on a 450MHz G3. Any suggestions on how to solve this problem would be greatly appreciated. Steve Means Naval Research Laboratory 4555 Overlook Avenue SW Code 7121 Washington, DC 20375 Email: means@wave.nrl.navy.mil Voice: 202-404-4653 Fax: 202-404-7813 -------------------------------- End of nih-image-d Digest V99 Issue #241 **************************************** From nih-image-request@io.ece.drexel.edu Thu Oct 21 13:21 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA07482; Thu, 21 Oct 1999 13:21:55 -0400 (EDT) Resent-Date: Thu, 21 Oct 1999 13:21:55 -0400 (EDT) Message-Id: <4.2.0.58.19991021115706.0095cc70@students.wisc.edu> X-Sender: kahowell@students.wisc.edu X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.0.58 Date: Thu, 21 Oct 1999 12:03:10 -0500 To: nih-image@io.ece.drexel.edu From: Kimberly A Howell Subject: FFT and Power Spectrum Mime-Version: 1.0 Resent-Message-ID: <"87A2Y3.0.-I1.CWq3u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1858 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii"; format=flowed Content-Length: 776 Status: O Hello, Does anyone know how to change the scaling in version 1.61 for the power spectrum? I noticed in Arlo's thesis that in version 1.25 there was a pull down menu called FFT Settings that allowed you to choose linear, log, or nth root scaling of the power spectrum. I haven't been able to find such a menu or place to change that option for version 1.61. I have a sneaking suspicion that the program is just doing log scaling and for the particular application of the power spectrum I want, I need linear scaling. Does anyone know about that? Please respond directly to me at kahowell@students.wisc.edu. Thanks in advance! Sincerely, Kim Howell Kimberly Howell, M.S. University of Wisconsin - Madison Department of Medical Physics Phone: (608) 263-9257 From nih-image-d-request@io.ece.drexel.edu Fri Oct 22 06:00 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA20147; Fri, 22 Oct 1999 06:00:54 -0400 (EDT) Date: Fri, 22 Oct 1999 06:00:54 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910221000.GAA20147@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #242 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/242 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 2601 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 242 Today's Topics: FFT and Power Spectrum [ Kimberly A Howell To: nih-image@io.ece.drexel.edu Subject: FFT and Power Spectrum Message-Id: <4.2.0.58.19991021115706.0095cc70@students.wisc.edu> Content-Type: text/plain; charset="us-ascii"; format=flowed Hello, Does anyone know how to change the scaling in version 1.61 for the power spectrum? I noticed in Arlo's thesis that in version 1.25 there was a pull down menu called FFT Settings that allowed you to choose linear, log, or nth root scaling of the power spectrum. I haven't been able to find such a menu or place to change that option for version 1.61. I have a sneaking suspicion that the program is just doing log scaling and for the particular application of the power spectrum I want, I need linear scaling. Does anyone know about that? Please respond directly to me at kahowell@students.wisc.edu. Thanks in advance! Sincerely, Kim Howell Kimberly Howell, M.S. University of Wisconsin - Madison Department of Medical Physics Phone: (608) 263-9257 ------------------------------ Date: Fri, 22 Oct 1999 11:51:40 +0200 From: Norbert Vischer To: nih-image@io.ece.drexel.edu Subject: Re: SetOptions('Minor') problem Message-Id: Content-Type: text/plain; charset="us-ascii" >SetOptions('X-Y Center,Major,Angle,Area'); - Works. >SetOptions('X-Y Center,Major,Minor,Angle,Area'); - Does not work. >Steve Means >Email: means@wave.nrl.navy.mil This is due to a bug in the CodeWarrior compiler, which I have already reported to Metroworks and Wayne. 'Minor' is exactly the 8th measurement column, and the compiler wrongly treats the 8th bit as sign bit and propagates it to the higher bits if it is set. In Object-Image the code is already changed so that "SetOptions" works as expected. Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html -------------------------------- End of nih-image-d Digest V99 Issue #242 **************************************** From nih-image-request@io.ece.drexel.edu Fri Oct 22 06:07 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA21458; Fri, 22 Oct 1999 06:07:11 -0400 (EDT) Resent-Date: Fri, 22 Oct 1999 06:07:11 -0400 (EDT) X-Authentication-Warning: mail.bio.uva.nl: Host gold.bio.uva.nl [145.18.160.48] claimed to be [145.18.160.48] Message-Id: In-Reply-To: Mime-Version: 1.0 Date: Fri, 22 Oct 1999 11:51:40 +0200 To: nih-image@io.ece.drexel.edu From: Norbert Vischer Subject: Re: SetOptions('Minor') problem Resent-Message-ID: <"E_C8K.0.6O4.NC34u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1859 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 834 Status: O >SetOptions('X-Y Center,Major,Angle,Area'); - Works. >SetOptions('X-Y Center,Major,Minor,Angle,Area'); - Does not work. >Steve Means >Email: means@wave.nrl.navy.mil This is due to a bug in the CodeWarrior compiler, which I have already reported to Metroworks and Wayne. 'Minor' is exactly the 8th measurement column, and the compiler wrongly treats the 8th bit as sign bit and propagates it to the higher bits if it is set. In Object-Image the code is already changed so that "SetOptions" works as expected. Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html From nih-image-request@io.ece.drexel.edu Fri Oct 22 09:20 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA23644; Fri, 22 Oct 1999 09:20:36 -0400 (EDT) Resent-Date: Fri, 22 Oct 1999 09:20:36 -0400 (EDT) From: "Steven Hudson" To: "Kimberly A Howell" Cc: Subject: RE: FFT and Power Spectrum Date: Fri, 22 Oct 1999 09:01:33 -0400 Message-ID: MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) X-Mimeole: Produced By Microsoft MimeOLE V5.00.2314.1300 Importance: Normal In-Reply-To: <4.2.0.58.19991021115706.0095cc70@students.wisc.edu> Resent-Message-ID: <"hdaUh2.0.Xy4.Z164u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1860 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="iso-8859-1" Content-Length: 1537 Status: O Yes, I think that it does log scaling, and to complicate matters, it seems that the base of the logarithm is selected automatically by the program. This is useful for display purposes, but not for quantitative analysis of the FFT. It would be nice if the base that is used by the program was output to the user. Steve Hudson Associate Professor of Macromolecular Science and Engineering 337 Kent Smith Building Case Western Reserve University 10900 Euclid Av. Cleveland, OH 44106-7202 216-368-6373 216-368-4202 fax > -----Original Message----- > From: Kimberly A Howell [mailto:kahowell@students.wisc.edu] > Sent: Thursday, October 21, 1999 1:03 PM > To: nih-image@io.ece.drexel.edu > Subject: FFT and Power Spectrum > > > Hello, > > Does anyone know how to change the scaling in version 1.61 for the power > spectrum? I noticed in Arlo's thesis that in version 1.25 there > was a pull > down menu called FFT Settings that allowed you to choose linear, log, or > nth root scaling of the power spectrum. I haven't been able to > find such a > menu or place to change that option for version 1.61. I have a sneaking > suspicion that the program is just doing log scaling and for the > particular > application of the power spectrum I want, I need linear scaling. Does > anyone know about that? Please respond directly to me at > kahowell@students.wisc.edu. Thanks in advance! > > Sincerely, > Kim Howell > > Kimberly Howell, M.S. > University of Wisconsin - Madison > Department of Medical Physics > Phone: (608) 263-9257 > From nih-image-d-request@io.ece.drexel.edu Sat Oct 23 05:24 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA23022; Sat, 23 Oct 1999 05:24:23 -0400 (EDT) Date: Sat, 23 Oct 1999 05:24:23 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910230924.FAA23022@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #243 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/243 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 6280 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 243 Today's Topics: RE: FFT and Power Spectrum [ "Steven Hudson" ] ADMIN: list commands [ nih-image-owner@io.ece.drexel.edu ] ------------------------------ Date: Fri, 22 Oct 1999 09:01:33 -0400 From: "Steven Hudson" To: "Kimberly A Howell" Cc: Subject: RE: FFT and Power Spectrum Message-ID: Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Yes, I think that it does log scaling, and to complicate matters, it seems that the base of the logarithm is selected automatically by the program. This is useful for display purposes, but not for quantitative analysis of the FFT. It would be nice if the base that is used by the program was output to the user. Steve Hudson Associate Professor of Macromolecular Science and Engineering 337 Kent Smith Building Case Western Reserve University 10900 Euclid Av. Cleveland, OH 44106-7202 216-368-6373 216-368-4202 fax > -----Original Message----- > From: Kimberly A Howell [mailto:kahowell@students.wisc.edu] > Sent: Thursday, October 21, 1999 1:03 PM > To: nih-image@io.ece.drexel.edu > Subject: FFT and Power Spectrum > > > Hello, > > Does anyone know how to change the scaling in version 1.61 for the power > spectrum? I noticed in Arlo's thesis that in version 1.25 there > was a pull > down menu called FFT Settings that allowed you to choose linear, log, or > nth root scaling of the power spectrum. I haven't been able to > find such a > menu or place to change that option for version 1.61. I have a sneaking > suspicion that the program is just doing log scaling and for the > particular > application of the power spectrum I want, I need linear scaling. Does > anyone know about that? Please respond directly to me at > kahowell@students.wisc.edu. Thanks in advance! > > Sincerely, > Kim Howell > > Kimberly Howell, M.S. > University of Wisconsin - Madison > Department of Medical Physics > Phone: (608) 263-9257 > ------------------------------ Date: Sat, 23 Oct 1999 05:05:01 -0400 (EDT) From: nih-image-owner@io.ece.drexel.edu To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Message-Id: <199910230905.FAA18889@io.ece.drexel.edu> NIH-image Mailing List Help --------------------------- nih-image-d-request@biomed.drexel.edu - Aministration for Digest nih-image-request@biomed.drexel.edu - Administation for List ********************************************************* * DO NOT SEND ADMINISTRATIVE COMMANDS TO THE LIST * ********************************************************* nih-image@biomed.drexel.edu - Sends mail to everyone on the list Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- -------------------------------- End of nih-image-d Digest V99 Issue #243 **************************************** From nih-image-request@io.ece.drexel.edu Sat Oct 23 05:30 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA24178; Sat, 23 Oct 1999 05:30:21 -0400 (EDT) Resent-Date: Sat, 23 Oct 1999 05:30:21 -0400 (EDT) Date: Sat, 23 Oct 1999 05:05:01 -0400 (EDT) From: nih-image-owner@io.ece.drexel.edu Message-Id: <199910230905.FAA18889@io.ece.drexel.edu> To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Resent-Message-ID: <"MgxUO2.0.Nd4._eN4u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1861 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 3743 Status: O NIH-image Mailing List Help --------------------------- nih-image-d-request@biomed.drexel.edu - Aministration for Digest nih-image-request@biomed.drexel.edu - Administation for List ********************************************************* * DO NOT SEND ADMINISTRATIVE COMMANDS TO THE LIST * ********************************************************* nih-image@biomed.drexel.edu - Sends mail to everyone on the list Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- From nih-image-request@io.ece.drexel.edu Mon Oct 25 16:15 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA07201; Mon, 25 Oct 1999 16:15:31 -0400 (EDT) Resent-Date: Mon, 25 Oct 1999 16:15:31 -0400 (EDT) Message-ID: <3814B795.975C59DD@umb.edu> Date: Mon, 25 Oct 1999 16:03:34 -0400 From: "Craig McClain" Reply-To: craig.mcclain@umb.edu X-Mailer: Mozilla 4.51 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: analyzing shape Resent-Message-ID: <"S3WFo2.0.ih.iJB5u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1862 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: multipart/mixed; boundary="------------014F54AAB75A44661C3818E0" Content-Length: 1210 Status: O This is a multi-part message in MIME format. --------------014F54AAB75A44661C3818E0 Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit I need help writing a macro that will take a 100 evenly space points along the outline of the shell. The macro must return to me the x,y coordinates of these points. In particular I want to obtain 100 points off the outline of a snail shell for eigenshape analysis. The macro would obtain these points of a scanned image of the shell itlself. thanks Craig Robert McClain --------------014F54AAB75A44661C3818E0 Content-Type: text/x-vcard; charset=us-ascii; name="craig.mcclain.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Craig McClain Content-Disposition: attachment; filename="craig.mcclain.vcf" begin:vcard n:McClain;Craig Robert tel;work:617-287-6651 x-mozilla-html:FALSE url:www.geocities.com/ResearchTriangle/Campus/1371 org:University of Mass., Boston;Department of Biology adr:;;100 Morrissey Blvd.;Boston;MA;02125;USA version:2.1 email;internet:craig.mcclain@umb.edu x-mozilla-cpt:;3 fn:Craig Robert McClain end:vcard --------------014F54AAB75A44661C3818E0-- From nih-image-request@io.ece.drexel.edu Mon Oct 25 21:57 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id VAA05080; Mon, 25 Oct 1999 21:57:40 -0400 (EDT) Resent-Date: Mon, 25 Oct 1999 21:57:40 -0400 (EDT) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: craig.mcclain@umb.edu, nih-image@io.ece.drexel.edu Date: Tue, 26 Oct 1999 11:50:59 +1000 Subject: Re: analyzing shape boundary="------------014F54AAB75A44661 Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <2A0017F5C30@rna.bio.mq.edu.au> Resent-Message-ID: <"6fpdq1.0.Mf.tUG5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1863 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 1720 Status: O >Date: Mon, 25 Oct 1999 16:03:34 -0400 >From: "Craig McClain" >To: nih-image@io.ece.drexel.edu >Subject: analyzing shape > >I need help writing a macro that will take a 100 evenly space points >along the outline of the shell. The macro must return to me the x,y >coordinates of these points. In particular I want to obtain 100 points >off the outline of a snail shell for eigenshape analysis. The macro >would obtain these points of a scanned image of the shell itlself. > >thanks >Craig Robert McClain Craig, This is straightforward (in principle) using Image. Create a selection (Region of Interest(ROI)) using eg thresholding and autoOutline then "nCoordinates Returns the number of XY coordinates used to define the current selection. The coordinates are stored in the xCoordinates[] and yCoordinates[] built-in arrays." Does the spacing need to be even by subtended angle at center of mass or by perimeter increment? In either case, image macro language is adequate to allow efficient evaluation. The degree of automation obtained will be limited by image qualities and effectiveness of thresholding in allowing automated segmentation. Object-Image has some advantages over NIH-Image in this application. see rLeft[n],rTop[n] Some years ago, I collaborated with Mick Howland to develope Fourier analysis of boundaries for shape analysis of fish otoliths in NIH-Image. Can you be more explicit as to what help you need? Sample scanned images by direct email would aid offlist discussion. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Tue Oct 26 06:40 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA05224; Tue, 26 Oct 1999 06:40:30 -0400 (EDT) Resent-Date: Tue, 26 Oct 1999 06:40:30 -0400 (EDT) Message-ID: <381582B4.106F@pet.hw.ac.uk> Date: Tue, 26 Oct 1999 11:30:12 +0100 From: katriona lightbody Reply-To: kl@pet.hw.ac.uk X-Mailer: Mozilla 3.04 (WinNT; I) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: perimeter too long Content-Transfer-Encoding: 7bit Resent-Message-ID: <"U-SlH.0.OW.H5O5u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1864 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 839 Status: O Hi I have a problem with a macro I have written and I was hoping that someone may be able to help. My macro loads a number of images, in this case 3, and applies the AnalyzeParticles function to each in turn. From previous analysis image1 contains 2998 particles, image2 3246 and image3 1449 particles. MaxMeasurements is set to 4095 and I am using ResetCounter before each call to AnalyseParticles, which is setting rCount to 0. However when my macro reaches the 3rd image and calls AnalyzeParticles, I get the 'Perimeter too long' error message at particle number 885. There is plenty of memory left and I don't have a clue what's going on. At this point the text within the package tends to take on a strange appearance and I have to quit and reload it. Any suggestions would be much appreciated. Thanks. Katriona Lightbody. From nih-image-d-request@io.ece.drexel.edu Tue Oct 26 17:38 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id RAA28678; Tue, 26 Oct 1999 17:38:57 -0400 (EDT) Date: Tue, 26 Oct 1999 17:38:57 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910262138.RAA28678@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #244 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/244 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 6839 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 244 Today's Topics: analyzing shape [ "Craig McClain" To: nih-image@io.ece.drexel.edu Subject: analyzing shape Message-ID: <3814B795.975C59DD@umb.edu> Content-Type: multipart/mixed; boundary="------------014F54AAB75A44661C3818E0" This is a multi-part message in MIME format. --------------014F54AAB75A44661C3818E0 Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit I need help writing a macro that will take a 100 evenly space points along the outline of the shell. The macro must return to me the x,y coordinates of these points. In particular I want to obtain 100 points off the outline of a snail shell for eigenshape analysis. The macro would obtain these points of a scanned image of the shell itlself. thanks Craig Robert McClain --------------014F54AAB75A44661C3818E0 Content-Type: text/x-vcard; charset=us-ascii; name="craig.mcclain.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Craig McClain Content-Disposition: attachment; filename="craig.mcclain.vcf" begin:vcard n:McClain;Craig Robert tel;work:617-287-6651 x-mozilla-html:FALSE url:www.geocities.com/ResearchTriangle/Campus/1371 org:University of Mass., Boston;Department of Biology adr:;;100 Morrissey Blvd.;Boston;MA;02125;USA version:2.1 email;internet:craig.mcclain@umb.edu x-mozilla-cpt:;3 fn:Craig Robert McClain end:vcard --------------014F54AAB75A44661C3818E0-- ------------------------------ Date: Tue, 26 Oct 1999 11:50:59 +1000 From: GJOSS@rna.bio.mq.edu.au To: craig.mcclain@umb.edu, nih-image@io.ece.drexel.edu Subject: Re: analyzing shape boundary="------------014F54AAB75A44661 Message-ID: <2A0017F5C30@rna.bio.mq.edu.au> >Date: Mon, 25 Oct 1999 16:03:34 -0400 >From: "Craig McClain" >To: nih-image@io.ece.drexel.edu >Subject: analyzing shape > >I need help writing a macro that will take a 100 evenly space points >along the outline of the shell. The macro must return to me the x,y >coordinates of these points. In particular I want to obtain 100 points >off the outline of a snail shell for eigenshape analysis. The macro >would obtain these points of a scanned image of the shell itlself. > >thanks >Craig Robert McClain Craig, This is straightforward (in principle) using Image. Create a selection (Region of Interest(ROI)) using eg thresholding and autoOutline then "nCoordinates Returns the number of XY coordinates used to define the current selection. The coordinates are stored in the xCoordinates[] and yCoordinates[] built-in arrays." Does the spacing need to be even by subtended angle at center of mass or by perimeter increment? In either case, image macro language is adequate to allow efficient evaluation. The degree of automation obtained will be limited by image qualities and effectiveness of thresholding in allowing automated segmentation. Object-Image has some advantages over NIH-Image in this application. see rLeft[n],rTop[n] Some years ago, I collaborated with Mick Howland to develope Fourier analysis of boundaries for shape analysis of fish otoliths in NIH-Image. Can you be more explicit as to what help you need? Sample scanned images by direct email would aid offlist discussion. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Tue, 26 Oct 1999 11:30:12 +0100 From: katriona lightbody To: nih-image@io.ece.drexel.edu Subject: perimeter too long Message-ID: <381582B4.106F@pet.hw.ac.uk> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi I have a problem with a macro I have written and I was hoping that someone may be able to help. My macro loads a number of images, in this case 3, and applies the AnalyzeParticles function to each in turn. From previous analysis image1 contains 2998 particles, image2 3246 and image3 1449 particles. MaxMeasurements is set to 4095 and I am using ResetCounter before each call to AnalyseParticles, which is setting rCount to 0. However when my macro reaches the 3rd image and calls AnalyzeParticles, I get the 'Perimeter too long' error message at particle number 885. There is plenty of memory left and I don't have a clue what's going on. At this point the text within the package tends to take on a strange appearance and I have to quit and reload it. Any suggestions would be much appreciated. Thanks. Katriona Lightbody. ------------------------------ Date: Tue, 26 Oct 1999 17:23:19 -0400 (EDT) From: Siamak Ardekani To: nih-image@io.ece.drexel.edu Subject: Moments based alignment Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, I am trying to write a macro to register sections using invariant moments. I have been using measurement command to compute the centroid and the best fit ellipse angle. The problem arises when there are multiple ROIs. If I use "Analyze Particles", each ROI would be considered separately. The only way to group the ROIs is to use wand tool and shift keys which in this case the macro won't be automatic. Any suggestion with this regards would be greatly appreciated. Thanks, Siamak Ardekani *************************************************************************** *Imaging & Computer Vision Center * *Drexel University E-mail:ardekani@io.ece.drexel.edu * *32nd and Chestnut Street * *Philadelphia, PA 19104 * * * *Home page: http://www.pages.drexel.edu/grad/sa26/ * * * *Tel:(215) 895-2279 * *Fax:(215) 895-4987 * *************************************************************************** -------------------------------- End of nih-image-d Digest V99 Issue #244 **************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 26 17:42 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id RAA29525; Tue, 26 Oct 1999 17:42:51 -0400 (EDT) Resent-Date: Tue, 26 Oct 1999 17:42:51 -0400 (EDT) Date: Tue, 26 Oct 1999 17:23:19 -0400 (EDT) From: Siamak Ardekani To: nih-image@io.ece.drexel.edu Subject: Moments based alignment Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"43F8-3.0.B96.9lX5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1865 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1353 Status: O Hello, I am trying to write a macro to register sections using invariant moments. I have been using measurement command to compute the centroid and the best fit ellipse angle. The problem arises when there are multiple ROIs. If I use "Analyze Particles", each ROI would be considered separately. The only way to group the ROIs is to use wand tool and shift keys which in this case the macro won't be automatic. Any suggestion with this regards would be greatly appreciated. Thanks, Siamak Ardekani *************************************************************************** *Imaging & Computer Vision Center * *Drexel University E-mail:ardekani@io.ece.drexel.edu * *32nd and Chestnut Street * *Philadelphia, PA 19104 * * * *Home page: http://www.pages.drexel.edu/grad/sa26/ * * * *Tel:(215) 895-2279 * *Fax:(215) 895-4987 * *************************************************************************** From nih-image-request@io.ece.drexel.edu Tue Oct 26 18:26 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA07693; Tue, 26 Oct 1999 18:26:45 -0400 (EDT) Resent-Date: Tue, 26 Oct 1999 18:26:45 -0400 (EDT) Message-Id: Mime-Version: 1.0 Date: Tue, 26 Oct 1999 15:18:47 -0700 To: nih-image@io.ece.drexel.edu From: Beth Bell Subject: statistical analysis Resent-Message-ID: <"E0Nid3.0.z71.NRY5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1866 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 864 Status: O Hello All, This is a question more about statistics than imaging per se. I collect images of a set number of sections (anterior to posterior) representing an hypothalamic nucleus, carefully matching anterior and posterior landmarks between animls. When I compare animals, what are the accepted methods? I have previously summed the values for each animal, getting a single number for each brain, or dividing it into anterior and posterior sections and summing accordingly. I've recently been told that using analysis of variance with repeated measures is also an acceptable method, since the measurement of the optical density is repeated through space (rather than time, which is how I usually think of it). Does anyone on this list use such a method of comparisons? Thank you, BETH BELL ___________ Procrastination KILLS Peace of Mind ___________ From nih-image-request@io.ece.drexel.edu Tue Oct 26 18:45 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA11161; Tue, 26 Oct 1999 18:45:48 -0400 (EDT) Resent-Date: Tue, 26 Oct 1999 18:45:48 -0400 (EDT) Date: Tue, 26 Oct 1999 18:25:46 -0400 (EDT) From: Brian Demczyk To: nih-image@io.ece.drexel.edu Subject: Scion AG5 NuBus Grabber Board Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"4o4-M2.0.Jy1.rhY5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1867 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 114 Status: O I have a Scion AG5 NuBus Grabber board, barely used that I can no longer use. Anyone interested? B. G. Demczyk From nih-image-request@io.ece.drexel.edu Tue Oct 26 19:34 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA19674; Tue, 26 Oct 1999 19:34:32 -0400 (EDT) Resent-Date: Tue, 26 Oct 1999 19:34:32 -0400 (EDT) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: kl@pet.hw.ac.uk, nih-image@io.ece.drexel.edu Date: Wed, 27 Oct 1999 9:25:46 +1000 Subject: Re: perimeter too long Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <2B597A619B7@rna.bio.mq.edu.au> Resent-Message-ID: <"ehy622.0.M94.mSZ5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1868 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 2280 Status: O >Date: Tue, 26 Oct 1999 11:30:12 +0100 >From: katriona lightbody >To: nih-image@io.ece.drexel.edu >Subject: perimeter too long > >Hi > >I have a problem with a macro I have written and I was hoping that >someone may be able to help. > >My macro loads a number of images, in this case 3, and applies the >AnalyzeParticles function to each in turn. From previous analysis >image1 contains 2998 particles, image2 3246 and image3 1449 particles. >MaxMeasurements is set to 4095 and I am using ResetCounter before each >call to AnalyseParticles, which is setting rCount to 0. However when >my >macro reaches the 3rd image and calls AnalyzeParticles, I get the >'Perimeter too long' error message at particle number 885. There is >plenty of memory left and I don't have a clue what's going on. At this >point the text within the package tends to take on a strange appearance >and I have to quit and reload it. > >Any suggestions would be much appreciated. > >Thanks. > >Katriona Lightbody. > 'Perimeter too long' error message at particle number 885 means just that Image has found a 'particle' whose perimeter exceeded the max 10,000 coordinate pairs allowed ie a large 'particle' with a complex shape. This is usually the result of incorrect theshold or else an artifact on image boundary which causes the intended particles to be linked together with many others (around the boundary). It has nothing to do with the results table MaxMeasurements or available memory. You should be able to reproduce the problem manually using the 'wand' tool by attempting to select the offending giant 'particle' with the same threshold or densitySlice. As you suggest that image3 worked before OK, the macro error may be associated with threshold or densitySlice so that you can insert GetThresholds(lower,upper);showMessage(lower,upper); in the macro to aid debug I would also recommend that you use Object-Image which has excellent debug facilities. Object-Image http://simon.bio.uva.nl/object-image.html by Norbert Vischer is an extention of Wayne Rasband's NIH Image. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Tue Oct 26 20:41 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id UAA02189; Tue, 26 Oct 1999 20:41:37 -0400 (EDT) Resent-Date: Tue, 26 Oct 1999 20:41:37 -0400 (EDT) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: ardekani@io.ece.drexel.edu, nih-image@io.ece.drexel.edu Date: Wed, 27 Oct 1999 10:31:43 +1000 Subject: Re: Moments based alignment Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <2B6B1412251@rna.bio.mq.edu.au> Resent-Message-ID: <"69rN91.0.s27.QQa5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1869 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 1583 Status: O >Date: Tue, 26 Oct 1999 17:23:19 -0400 (EDT) >From: Siamak Ardekani >To: nih-image@io.ece.drexel.edu >Subject: Moments based alignment >Hello, > > I am trying to write a macro to register sections using invariant >moments. I have been using measurement command to compute the centroid >and the best fit ellipse angle. The problem arises when there are >multiple ROIs. If I use "Analyze Particles", each ROI would be >considered separately. The only way to group the ROIs is to use wand >tool and shift keys which in this case the macro won't be automatic. > >Any suggestion with this regards would be greatly appreciated. >Thanks, >Siamak Ardekani > AnalyzeParticles returns the centroid and area (mass equivalent) of the binary particles. To calculate the centroid and then moments, angle of the 'combined' particle: procedure moments;var i,mx,my,a:real;begin mx:=0;my:=0;a:=0; for i:=1 to rcount do begin a:=a+rArea[i]; mx:=mx+rArea[i]*rX[i]; my:=my+rArea[i]*rY[i]; end; x:=mx/a;y:=my/a;{combined centroid at x,y} mx:=0;my:=0; for i:=1 to rcount do begin mx:=mx+rArea[i]*abs(rX[i]-x); my:=my+rArea[i]*abs(rX[i]-y); end; if my>0 then angle:=arctan(mx/my) else angle:=2*arctan(1);{radians} angle:=angle/arctan(1)*45;{degrees} end macro'moments';var x,y,angle:real;begin moments;showMessage(x,y,angle);end x,y,angle should be all you need for registration. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Wed Oct 27 01:30 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id BAA27736; Wed, 27 Oct 1999 01:30:27 -0400 (EDT) Resent-Date: Wed, 27 Oct 1999 01:30:27 -0400 (EDT) Message-ID: <000701bf203a$71acb840$990f3c86@biologie.uniulm.de> From: "Wilko Ahlrichs" To: References: <199910262123.RAA25235@io.ece.drexel.edu> Date: Wed, 27 Oct 1999 07:16:37 +0200 MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2014.211 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2014.211 Resent-Message-ID: <"m50BL3.0.g36.5he5u"@io> Resent-From: nih-image@io.ece.drexel.edu Subject: Unidentified subject! Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1870 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="iso-8859-1" Content-Length: 13 Status: O unsubscribe From nih-image-request@io.ece.drexel.edu Wed Oct 27 02:33 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id CAA10442; Wed, 27 Oct 1999 02:33:48 -0400 (EDT) Resent-Date: Wed, 27 Oct 1999 02:33:48 -0400 (EDT) X-Sender: ajonker.public.amc@reddwarf.amc.uva.nl Message-Id: In-Reply-To: <199910262126.RAA26109@io.ece.drexel.edu> Mime-Version: 1.0 Date: Wed, 27 Oct 1999 08:14:03 +0200 To: nih-image@io.ece.drexel.edu From: Ard Jonker Subject: Re: analyzing shape Resent-Message-ID: <"lSbmU1.0.9l1.zZf5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1872 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1726 Status: O >I need help writing a macro that will take a 100 evenly space points >along the outline of the shell. The macro must return to me the x,y >coordinates of these points. In particular I want to obtain 100 points >off the outline of a snail shell for eigenshape analysis. The macro >would obtain these points of a scanned image of the shell itlself. You can use ObjectImage to obtain the ROI coordinates (or Object Coordinates, once you have converted the ROI to an object). Objects can be stored nondestructively with the image. If you obtain m (m>100) coordinate pairs, you pick the 1.. int(m/100) coordinates. You might be fooled there, however, as you might miss strong curvatures if the shape changes in m...m+n, where int((n+m)/100)-int(m/100) is bigger than 1. I have devised several macros to smooth a ROI, keeping the shape as much intact as possible. Regards, ARd >thanks >Craig Robert McClain > >Content-Type: text/x-vcard; charset=us-ascii; > name="craig.mcclain.vcf" >Content-Transfer-Encoding: 7bit >Content-Description: Card for Craig McClain >Content-Disposition: attachment; > filename="craig.mcclain.vcf" > >Attachment converted: G3 B/W data disk:craig.mcclain.vcf (TEXT/ttxt) >(00009A99) Why are you attaching files that wind up on my harddisk? In the past few weeks a few dozen of these things land here and I keep shoveling this stuff out. Quite annoying. dr A.Jonker |Academic Medical Centre Dep of Cell Biology and Histology |Meibergdreef 15 Faculty of Medicine |1105 AZ Amsterdam University of Amsterdam |The Netherlands tel +31 20 566 6804 |fax +31 20 697 4156 PGP fingerprint B263 BEAF 48E2 FE6E 2525 B8C3 EBF9 7685 9A1E 8A3A From nih-image-request@io.ece.drexel.edu Wed Oct 27 02:34 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id CAA10689; Wed, 27 Oct 1999 02:34:49 -0400 (EDT) Resent-Date: Wed, 27 Oct 1999 02:34:49 -0400 (EDT) X-Sender: ajonker.public.amc@reddwarf.amc.uva.nl Message-Id: In-Reply-To: <199910262126.RAA26109@io.ece.drexel.edu> Mime-Version: 1.0 Date: Wed, 27 Oct 1999 08:17:12 +0200 To: nih-image@io.ece.drexel.edu From: Ard Jonker Subject: Re: perimeter too long Resent-Message-ID: <"7pt771.0.Nk1.xZf5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1871 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1597 Status: O >My macro loads a number of images, in this case 3, and applies the >AnalyzeParticles function to each in turn. From previous analysis >image1 contains 2998 particles, image2 3246 and image3 1449 particles. >MaxMeasurements is set to 4095 and I am using ResetCounter before each >call to AnalyseParticles, which is setting rCount to 0. However when my >macro reaches the 3rd image and calls AnalyzeParticles, I get the >'Perimeter too long' error message at particle number 885. There is >plenty of memory left and I don't have a clue what's going on. At this >point the text within the package tends to take on a strange appearance >and I have to quit and reload it. > >Any suggestions would be much appreciated. Hi Katriona The error message indicates that particle number 885 has too long a perimeter. That is, either this particle is very large, stretching the perimeter beyond NIH's limits (10.000 coordinate pairs, I believe) or it is very irregularly shaped (ie. with many "fjords" into the object) which also elongates the perimeter. In the first case, you could scale your image to 50% and redo the measurement. In the second case, you can smooth the image to reduce the perimeter. Both solutions are not optimal, I realise Regards, Ard dr A.Jonker |Academic Medical Centre Dep of Cell Biology and Histology |Meibergdreef 15 Faculty of Medicine |1105 AZ Amsterdam University of Amsterdam |The Netherlands tel +31 20 566 6804 |fax +31 20 697 4156 PGP fingerprint B263 BEAF 48E2 FE6E 2525 B8C3 EBF9 7685 9A1E 8A3A From nih-image-d-request@io.ece.drexel.edu Wed Oct 27 10:26 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id KAA05472; Wed, 27 Oct 1999 10:26:41 -0400 (EDT) Date: Wed, 27 Oct 1999 10:26:41 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910271426.KAA05472@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #245 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/245 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 12609 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 245 Today's Topics: statistical analysis [ Beth Bell ] Scion AG5 NuBus Grabber Board [ Brian Demczyk ] Re: analyzing shape [ Ard Jonker ] Unidentified subject! [ Kardi Teknomo ] ------------------------------ Date: Tue, 26 Oct 1999 15:18:47 -0700 From: Beth Bell To: nih-image@io.ece.drexel.edu Subject: statistical analysis Message-Id: Content-Type: text/plain; charset="us-ascii" Hello All, This is a question more about statistics than imaging per se. I collect images of a set number of sections (anterior to posterior) representing an hypothalamic nucleus, carefully matching anterior and posterior landmarks between animls. When I compare animals, what are the accepted methods? I have previously summed the values for each animal, getting a single number for each brain, or dividing it into anterior and posterior sections and summing accordingly. I've recently been told that using analysis of variance with repeated measures is also an acceptable method, since the measurement of the optical density is repeated through space (rather than time, which is how I usually think of it). Does anyone on this list use such a method of comparisons? Thank you, BETH BELL ___________ Procrastination KILLS Peace of Mind ___________ ------------------------------ Date: Tue, 26 Oct 1999 18:25:46 -0400 (EDT) From: Brian Demczyk To: nih-image@io.ece.drexel.edu Subject: Scion AG5 NuBus Grabber Board Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII I have a Scion AG5 NuBus Grabber board, barely used that I can no longer use. Anyone interested? B. G. Demczyk ------------------------------ Date: Wed, 27 Oct 1999 9:25:46 +1000 From: GJOSS@rna.bio.mq.edu.au To: kl@pet.hw.ac.uk, nih-image@io.ece.drexel.edu Subject: Re: perimeter too long Message-ID: <2B597A619B7@rna.bio.mq.edu.au> >Date: Tue, 26 Oct 1999 11:30:12 +0100 >From: katriona lightbody >To: nih-image@io.ece.drexel.edu >Subject: perimeter too long > >Hi > >I have a problem with a macro I have written and I was hoping that >someone may be able to help. > >My macro loads a number of images, in this case 3, and applies the >AnalyzeParticles function to each in turn. From previous analysis >image1 contains 2998 particles, image2 3246 and image3 1449 particles. >MaxMeasurements is set to 4095 and I am using ResetCounter before each >call to AnalyseParticles, which is setting rCount to 0. However when >my >macro reaches the 3rd image and calls AnalyzeParticles, I get the >'Perimeter too long' error message at particle number 885. There is >plenty of memory left and I don't have a clue what's going on. At this >point the text within the package tends to take on a strange appearance >and I have to quit and reload it. > >Any suggestions would be much appreciated. > >Thanks. > >Katriona Lightbody. > 'Perimeter too long' error message at particle number 885 means just that Image has found a 'particle' whose perimeter exceeded the max 10,000 coordinate pairs allowed ie a large 'particle' with a complex shape. This is usually the result of incorrect theshold or else an artifact on image boundary which causes the intended particles to be linked together with many others (around the boundary). It has nothing to do with the results table MaxMeasurements or available memory. You should be able to reproduce the problem manually using the 'wand' tool by attempting to select the offending giant 'particle' with the same threshold or densitySlice. As you suggest that image3 worked before OK, the macro error may be associated with threshold or densitySlice so that you can insert GetThresholds(lower,upper);showMessage(lower,upper); in the macro to aid debug I would also recommend that you use Object-Image which has excellent debug facilities. Object-Image http://simon.bio.uva.nl/object-image.html by Norbert Vischer is an extention of Wayne Rasband's NIH Image. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Wed, 27 Oct 1999 10:31:43 +1000 From: GJOSS@rna.bio.mq.edu.au To: ardekani@io.ece.drexel.edu, nih-image@io.ece.drexel.edu Subject: Re: Moments based alignment Message-ID: <2B6B1412251@rna.bio.mq.edu.au> >Date: Tue, 26 Oct 1999 17:23:19 -0400 (EDT) >From: Siamak Ardekani >To: nih-image@io.ece.drexel.edu >Subject: Moments based alignment >Hello, > > I am trying to write a macro to register sections using invariant >moments. I have been using measurement command to compute the centroid >and the best fit ellipse angle. The problem arises when there are >multiple ROIs. If I use "Analyze Particles", each ROI would be >considered separately. The only way to group the ROIs is to use wand >tool and shift keys which in this case the macro won't be automatic. > >Any suggestion with this regards would be greatly appreciated. >Thanks, >Siamak Ardekani > AnalyzeParticles returns the centroid and area (mass equivalent) of the binary particles. To calculate the centroid and then moments, angle of the 'combined' particle: procedure moments;var i,mx,my,a:real;begin mx:=0;my:=0;a:=0; for i:=1 to rcount do begin a:=a+rArea[i]; mx:=mx+rArea[i]*rX[i]; my:=my+rArea[i]*rY[i]; end; x:=mx/a;y:=my/a;{combined centroid at x,y} mx:=0;my:=0; for i:=1 to rcount do begin mx:=mx+rArea[i]*abs(rX[i]-x); my:=my+rArea[i]*abs(rX[i]-y); end; if my>0 then angle:=arctan(mx/my) else angle:=2*arctan(1);{radians} angle:=angle/arctan(1)*45;{degrees} end macro'moments';var x,y,angle:real;begin moments;showMessage(x,y,angle);end x,y,angle should be all you need for registration. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Wed, 27 Oct 1999 07:16:37 +0200 From: "Wilko Ahlrichs" To: Subject: Unidentified subject! Message-ID: <000701bf203a$71acb840$990f3c86@biologie.uniulm.de> Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit unsubscribe ------------------------------ Date: Wed, 27 Oct 1999 08:17:12 +0200 From: Ard Jonker To: nih-image@io.ece.drexel.edu Subject: Re: perimeter too long Message-Id: Content-Type: text/plain; charset="us-ascii" >My macro loads a number of images, in this case 3, and applies the >AnalyzeParticles function to each in turn. From previous analysis >image1 contains 2998 particles, image2 3246 and image3 1449 particles. >MaxMeasurements is set to 4095 and I am using ResetCounter before each >call to AnalyseParticles, which is setting rCount to 0. However when my >macro reaches the 3rd image and calls AnalyzeParticles, I get the >'Perimeter too long' error message at particle number 885. There is >plenty of memory left and I don't have a clue what's going on. At this >point the text within the package tends to take on a strange appearance >and I have to quit and reload it. > >Any suggestions would be much appreciated. Hi Katriona The error message indicates that particle number 885 has too long a perimeter. That is, either this particle is very large, stretching the perimeter beyond NIH's limits (10.000 coordinate pairs, I believe) or it is very irregularly shaped (ie. with many "fjords" into the object) which also elongates the perimeter. In the first case, you could scale your image to 50% and redo the measurement. In the second case, you can smooth the image to reduce the perimeter. Both solutions are not optimal, I realise Regards, Ard dr A.Jonker |Academic Medical Centre Dep of Cell Biology and Histology |Meibergdreef 15 Faculty of Medicine |1105 AZ Amsterdam University of Amsterdam |The Netherlands tel +31 20 566 6804 |fax +31 20 697 4156 PGP fingerprint B263 BEAF 48E2 FE6E 2525 B8C3 EBF9 7685 9A1E 8A3A ------------------------------ Date: Wed, 27 Oct 1999 08:14:03 +0200 From: Ard Jonker To: nih-image@io.ece.drexel.edu Subject: Re: analyzing shape Message-Id: Content-Type: text/plain; charset="us-ascii" >I need help writing a macro that will take a 100 evenly space points >along the outline of the shell. The macro must return to me the x,y >coordinates of these points. In particular I want to obtain 100 points >off the outline of a snail shell for eigenshape analysis. The macro >would obtain these points of a scanned image of the shell itlself. You can use ObjectImage to obtain the ROI coordinates (or Object Coordinates, once you have converted the ROI to an object). Objects can be stored nondestructively with the image. If you obtain m (m>100) coordinate pairs, you pick the 1.. int(m/100) coordinates. You might be fooled there, however, as you might miss strong curvatures if the shape changes in m...m+n, where int((n+m)/100)-int(m/100) is bigger than 1. I have devised several macros to smooth a ROI, keeping the shape as much intact as possible. Regards, ARd >thanks >Craig Robert McClain > >Content-Type: text/x-vcard; charset=us-ascii; > name="craig.mcclain.vcf" >Content-Transfer-Encoding: 7bit >Content-Description: Card for Craig McClain >Content-Disposition: attachment; > filename="craig.mcclain.vcf" > >Attachment converted: G3 B/W data disk:craig.mcclain.vcf (TEXT/ttxt) >(00009A99) Why are you attaching files that wind up on my harddisk? In the past few weeks a few dozen of these things land here and I keep shoveling this stuff out. Quite annoying. dr A.Jonker |Academic Medical Centre Dep of Cell Biology and Histology |Meibergdreef 15 Faculty of Medicine |1105 AZ Amsterdam University of Amsterdam |The Netherlands tel +31 20 566 6804 |fax +31 20 697 4156 PGP fingerprint B263 BEAF 48E2 FE6E 2525 B8C3 EBF9 7685 9A1E 8A3A ------------------------------ Date: Wed, 27 Oct 1999 16:33:02 +0900 (JST) From: Kardi Teknomo To: nih-image-d@io.ece.drexel.edu Subject: Unidentified subject! Message-Id: <199910270733.QAA13631@plan.civil.tohoku.ac.jp> Content-Type: text/plain; charset="us-ascii" Dear imager, Is anyone know the formula to get the gray value of the image math (OR, XOR, and AND)? For example, gray value of Image A in location (x,y) is 120 and in image B in the same location (x,y) is 190, then using image math (Image-A XOR image-B) x 1 + 0 = 198. What is the formula to get 198 from 190 and 120 ? Another examples are in the tables below: image A image B (A XOR B)x1+0 (A OR B)x1+0 (A AND B)x1+0 120 190 198 254 56 120 80 40 120 80 100 150 242 246 104 It seems there is a cyclic formula? Is there any references for this kind of Set problem? Thanks in advance. Kardi Kardi Teknomo Road Transportation and Traffic Laboratory Graduate School of Information Science Tohoku University Japan http://www.plan.civil.tohoku.ac.jp/~kardi/ e-mail: kardi@plan.civil.tohoku.ac.jp ------------------------------ Date: Wed, 27 Oct 99 10:07:51 -0400 From: jared rifkin To: "nih-image users group" Subject: Re: Scion AG5 NuBus Grabber Board Message-ID: <266157F6530@qc1.qc.edu> Content-Type: text/plain; charset="US-ASCII" what are specs (resolution, port configuration, etc)? -------------------------------- End of nih-image-d Digest V99 Issue #245 **************************************** From nih-image-request@io.ece.drexel.edu Wed Oct 27 10:30 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id KAA06356; Wed, 27 Oct 1999 10:30:16 -0400 (EDT) Resent-Date: Wed, 27 Oct 1999 10:30:16 -0400 (EDT) Subject: Re: Scion AG5 NuBus Grabber Board Date: Wed, 27 Oct 99 10:07:51 -0400 x-sender: jlr$biol@qc1.qc.edu x-mailer: Claris Emailer 1.1 From: jared rifkin To: "nih-image users group" Mime-Version: 1.0 Message-ID: <266157F6530@qc1.qc.edu> Resent-Message-ID: <"smt131.0.UJ.PTm5u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1873 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 55 Status: O what are specs (resolution, port configuration, etc)? From nih-image-request@io.ece.drexel.edu Wed Oct 27 10:31 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id KAA06535; Wed, 27 Oct 1999 10:31:00 -0400 (EDT) Resent-Date: Wed, 27 Oct 1999 10:31:00 -0400 (EDT) Message-ID: <38170B99.9494C5CF@umb.edu> Date: Wed, 27 Oct 1999 10:26:35 -0400 From: "Craig McClain" Reply-To: craig.mcclain@umb.edu X-Mailer: Mozilla 4.51 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: statistical analysis References: Resent-Message-ID: <"N0ZLs1.0.xi.oZm5u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1874 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: multipart/mixed; boundary="------------A40B34C67FD7FED1ED6995D9" Content-Length: 1024 Status: O This is a multi-part message in MIME format. --------------A40B34C67FD7FED1ED6995D9 Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit I might be able to help you. Could you elaborate more on what exactly your data set is and what your hypothesis(es) are? What is it that you want to be able to say? -Craig Robert McClain --------------A40B34C67FD7FED1ED6995D9 Content-Type: text/x-vcard; charset=us-ascii; name="craig.mcclain.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Craig McClain Content-Disposition: attachment; filename="craig.mcclain.vcf" begin:vcard n:McClain;Craig Robert tel;work:617-287-6651 x-mozilla-html:FALSE url:www.geocities.com/ResearchTriangle/Campus/1371 org:University of Mass., Boston;Department of Biology adr:;;100 Morrissey Blvd.;Boston;MA;02125;USA version:2.1 email;internet:craig.mcclain@umb.edu x-mozilla-cpt:;3 fn:Craig Robert McClain end:vcard --------------A40B34C67FD7FED1ED6995D9-- From nih-image-request@io.ece.drexel.edu Wed Oct 27 23:54 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id XAA13299; Wed, 27 Oct 1999 23:54:02 -0400 (EDT) Resent-Date: Wed, 27 Oct 1999 23:54:02 -0400 (EDT) Message-ID: <3817C678.C67DE7ED@thurston.com> Date: Wed, 27 Oct 1999 20:43:55 -0700 From: Pat and Leslie Pringle Reply-To: lespat@thurston.com X-Mailer: Mozilla 4.05 (Macintosh; I; PPC) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #245 References: <199910271412.KAA02328@io.ece.drexel.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"BJJzA2.0.id2.kLy5u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1875 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Length: 953 Status: O To the list: Over the years of being a member of this list, I've noticed something quite amazing: the dedication with which certain people help their colleagues--by selflessly sharing their expertise, knowledge, creativity, and time. It's enormously impressive, and we should have some kind of award to at least acknowledge these folks for their dedication. Their contributions are a great gift to us all. I'd like to thank them all, and suggest we give them, at least ceremonially, a an award--we could call it the Golden Image Award (for lack of a better idea). Here are a few of the people who have helped me and others (my apologies to anyone not mentioned--this list is not exhaustive). Wayne Rasband John Russ Norbert Vischer Greg Joss Ard Jonker the Schuettes Mark Vivino others? I'm sure there are many more... Anyway, my thanks...you;'ve been an enormous help to us all. Please excuse me for mispellings. Pat Pringle Olympia, Washington USA From nih-image-d-request@io.ece.drexel.edu Thu Oct 28 05:30 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA11496; Thu, 28 Oct 1999 05:30:57 -0400 (EDT) Date: Thu, 28 Oct 1999 05:30:57 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910280930.FAA11496@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #246 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/246 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5176 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 246 Today's Topics: Re: statistical analysis [ "Craig McClain" ] ------------------------------ Date: Wed, 27 Oct 1999 10:26:35 -0400 From: "Craig McClain" To: nih-image@io.ece.drexel.edu Subject: Re: statistical analysis Message-ID: <38170B99.9494C5CF@umb.edu> Content-Type: multipart/mixed; boundary="------------A40B34C67FD7FED1ED6995D9" This is a multi-part message in MIME format. --------------A40B34C67FD7FED1ED6995D9 Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit I might be able to help you. Could you elaborate more on what exactly your data set is and what your hypothesis(es) are? What is it that you want to be able to say? -Craig Robert McClain --------------A40B34C67FD7FED1ED6995D9 Content-Type: text/x-vcard; charset=us-ascii; name="craig.mcclain.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Craig McClain Content-Disposition: attachment; filename="craig.mcclain.vcf" begin:vcard n:McClain;Craig Robert tel;work:617-287-6651 x-mozilla-html:FALSE url:www.geocities.com/ResearchTriangle/Campus/1371 org:University of Mass., Boston;Department of Biology adr:;;100 Morrissey Blvd.;Boston;MA;02125;USA version:2.1 email;internet:craig.mcclain@umb.edu x-mozilla-cpt:;3 fn:Craig Robert McClain end:vcard --------------A40B34C67FD7FED1ED6995D9-- ------------------------------ Date: Thu, 28 Oct 1999 10:37:48 +0900 (JST) From: Kardi Teknomo To: nih-image-d@io.ece.drexel.edu Subject: Set theory of gray value Message-Id: <199910280137.KAA18099@plan.civil.tohoku.ac.jp> Content-Type: text/plain; charset="us-ascii" Dear imager, Is anyone know the formula to get the gray value of the image math (OR, XOR, and AND)? For example, gray value of Image A in location (x,y) is 120 and in image B in the same location (x,y) is 190, then using image math (Image-A XOR image-B) x 1 + 0 = 198. What is the formula to get 198 from 190 and 120 ? Another examples are in the tables below: image A image B (A XOR B)x1+0 (A OR B)x1+0 (A AND B)x1+0 120 190 198 254 56 120 80 40 120 80 100 150 242 246 104 It seems there is a cyclic formula? Is there any references for this kind of Set problem? Thanks in advance. Kardi Teknomo Road Transportation and Traffic Laboratory Graduate School of Information Science Tohoku University Japan http://www.plan.civil.tohoku.ac.jp/~kardi/ e-mail: kardi@plan.civil.tohoku.ac.jp ------------------------------ Date: Wed, 27 Oct 1999 20:43:55 -0700 From: Pat and Leslie Pringle To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #245 Message-ID: <3817C678.C67DE7ED@thurston.com> Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit To the list: Over the years of being a member of this list, I've noticed something quite amazing: the dedication with which certain people help their colleagues--by selflessly sharing their expertise, knowledge, creativity, and time. It's enormously impressive, and we should have some kind of award to at least acknowledge these folks for their dedication. Their contributions are a great gift to us all. I'd like to thank them all, and suggest we give them, at least ceremonially, a an award--we could call it the Golden Image Award (for lack of a better idea). Here are a few of the people who have helped me and others (my apologies to anyone not mentioned--this list is not exhaustive). Wayne Rasband John Russ Norbert Vischer Greg Joss Ard Jonker the Schuettes Mark Vivino others? I'm sure there are many more... Anyway, my thanks...you;'ve been an enormous help to us all. Please excuse me for mispellings. Pat Pringle Olympia, Washington USA ------------------------------ Date: Thu, 28 Oct 1999 05:14:40 -0400 From: Lance Davidson To: nih-image@io.ece.drexel.edu Subject: Serial Control thru B&W G3 USB Message-ID: <38181400.CE1E15B8@virginia.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit We currently control a Uniblitz shutter with our reliable old Quadra 840AV through its standard serial port. However we want to move this shutter control to a B&W G3 that has no serial port. I've obtained a Keyspan USB to Serial converter but Image just doesn't seem to recognize the port. Does anyone have any suggestions? Thanks, Lance _________________________________ Lance Davidson University of Virginia Department of Biology Charlottesville, VA 22903 804-243-2596 -------------------------------- End of nih-image-d Digest V99 Issue #246 **************************************** From nih-image-request@io.ece.drexel.edu Thu Oct 28 05:34 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA12265; Thu, 28 Oct 1999 05:34:29 -0400 (EDT) Resent-Date: Thu, 28 Oct 1999 05:34:29 -0400 (EDT) Message-ID: <38181400.CE1E15B8@virginia.edu> Date: Thu, 28 Oct 1999 05:14:40 -0400 From: Lance Davidson Organization: Department of Biology X-Mailer: Mozilla 4.05 [en] (Win95; I) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Serial Control thru B&W G3 USB Content-Transfer-Encoding: 7bit Resent-Message-ID: <"CJzgP1.0.z52.-G16u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1876 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 474 Status: O We currently control a Uniblitz shutter with our reliable old Quadra 840AV through its standard serial port. However we want to move this shutter control to a B&W G3 that has no serial port. I've obtained a Keyspan USB to Serial converter but Image just doesn't seem to recognize the port. Does anyone have any suggestions? Thanks, Lance _________________________________ Lance Davidson University of Virginia Department of Biology Charlottesville, VA 22903 804-243-2596 From nih-image-request@io.ece.drexel.edu Thu Oct 28 09:06 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA17669; Thu, 28 Oct 1999 09:06:18 -0400 (EDT) Resent-Date: Thu, 28 Oct 1999 09:06:18 -0400 (EDT) User-Agent: Microsoft Outlook Express Macintosh Edition - 5.0 (1513) Date: Thu, 28 Oct 1999 08:49:06 +0000 Subject: Re: nih-image-d Digest V99 #245 From: Marcia Goldfarb To: Message-ID: In-Reply-To: <3817C678.C67DE7ED@thurston.com> Mime-version: 1.0 Content-transfer-encoding: 7bit Resent-Message-ID: <"LJIBe2.0.-V3.QO46u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1877 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="US-ASCII" Content-Length: 426 Status: O I want to second the email from Pat and Leslie Pringle. This list is so helpful. It is the first list I joined, and it spoiled me for subsequent newsgroups. I am amazed by the rude and crude responses in other scientific forums. Certain "Imagers" are really knowledgeable and willing to help, and everyone is professional Thank you Marcia Goldfarb Anatek-EP 17 Bishop St Portland, ME 04103 email: anatekep@maine.rr.com From nih-image-request@io.ece.drexel.edu Thu Oct 28 09:23 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA20671; Thu, 28 Oct 1999 09:23:09 -0400 (EDT) Resent-Date: Thu, 28 Oct 1999 09:23:09 -0400 (EDT) From: ort056@abdn.ac.uk Sender: ort056@abdn.ac.uk To: nih-image@io.ece.drexel.edu Subject: award Message-ID: Date: Thu, 28 Oct 1999 14:10:23 +0100 (BST) Priority: NORMAL X-Mailer: Simeon for Windows Version 4.0.9 X-Authentication: IMSP MIME-Version: 1.0 Resent-Message-ID: <"gDRAZ.0.mM4.zf46u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1878 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Length: 292 Status: O I agree, There can't be many subscribers who haven't been helped by the people nominated, it would be great to have a beeter way to thank them. ---------------------- Jennifer Gregory j.gregory@abdn.ac.uk Department of Orthopaedics Polwarth Building Foresterhill Aberdeen AB25 2QJ From nih-image-request@io.ece.drexel.edu Thu Oct 28 17:01 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id RAA02030; Thu, 28 Oct 1999 17:01:33 -0400 (EDT) Resent-Date: Thu, 28 Oct 1999 17:01:33 -0400 (EDT) Message-ID: From: "Berger, Jennifer" To: "'NIH Image List'" Subject: Macro Help Date: Thu, 28 Oct 1999 14:41:27 -0600 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2448.0) Resent-Message-ID: <"-VxW03.0.Nv6.9KB6u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1879 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="windows-1252" Content-Length: 416 Status: O I am wondering if it is possible (and how to go about doing so) to write a Macro for NIH Image. I have a Scion CG-7 Framegrabber. The problem that I am having is that I need to get my image into 8-Bit color before I can copy it to another program. As things work now I capture in 24-Bit Color, then go to RGB, then finally to 8-Bit. Is there a way to make this one step. Thanks in advance for any help Jennifer From nih-image-request@io.ece.drexel.edu Thu Oct 28 19:31 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA25917; Thu, 28 Oct 1999 19:31:16 -0400 (EDT) Resent-Date: Thu, 28 Oct 1999 19:31:16 -0400 (EDT) Mime-Version: 1.0 X-Sender: scm@128.250.6.196 Message-Id: In-Reply-To: References: Date: Fri, 29 Oct 1999 09:12:46 +1000 To: nih-image@io.ece.drexel.edu From: Steve Martin Subject: Re: award Resent-Message-ID: <"WaWG13.0.uU5.0YD6u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1880 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" ; format="flowed" Content-Length: 619 Status: O At 2:10 PM +0100 on 28/10/99, ort056@abdn.ac.uk wrote: > I agree, There can't be many subscribers who haven't been helped by > the people nominated, it would be great to have a beeter way to thank > them. perhaps it should be called the 'Stained Cup', in reference to all the histological users of NIH Image, or the 'Particular Trophy'. I would be willing to donate a small 3d virtual trophy and a java applet to display it. The annual award holder would be the only person licensed to display it on their web page. I nominate Mr Joss if the annual volume of macro code/pseudocode posted is a criterion. Steve From nih-image-request@io.ece.drexel.edu Thu Oct 28 23:04 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id XAA04706; Thu, 28 Oct 1999 23:04:15 -0400 (EDT) Resent-Date: Thu, 28 Oct 1999 23:04:15 -0400 (EDT) X-Sender: drmason@mail.interconnect.com.au Message-Id: In-Reply-To: References: Mime-Version: 1.0 Date: Fri, 29 Oct 1999 12:13:09 +1030 To: nih-image@io.ece.drexel.edu From: Doug Mason Subject: Re: award Resent-Message-ID: <"s_gM63.0.9H.jeG6u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1881 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 627 Status: O Steve Martin said: >I nominate Mr Joss if the annual volume of macro code/pseudocode >posted is a criterion. > I second the nomination. I'm sure many others would, too. To the 'annual volume' criterion we could add many other characteristics of Greg Joss's contributions (like insight into problem solving, facility with the macro language, speed of response, etc.). Thanks, Greg! Doug Mason Mason Geoscience Pty Ltd Petrological Services for the Exploration and Mining Industry email : drmason@interconnect.com.au post : PO Box 78, Glenside SA 5065, Australia phone : +61-8-83901507 fax : +61-8-83901194 From nih-image-d-request@io.ece.drexel.edu Fri Oct 29 00:48 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id AAA24700; Fri, 29 Oct 1999 00:48:01 -0400 (EDT) Date: Fri, 29 Oct 1999 00:48:01 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910290448.AAA24700@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #247 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/247 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5489 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 247 Today's Topics: Re: nih-image-d Digest V99 #245 [ Marcia Goldfarb To: Subject: Re: nih-image-d Digest V99 #245 Message-ID: Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit I want to second the email from Pat and Leslie Pringle. This list is so helpful. It is the first list I joined, and it spoiled me for subsequent newsgroups. I am amazed by the rude and crude responses in other scientific forums. Certain "Imagers" are really knowledgeable and willing to help, and everyone is professional Thank you Marcia Goldfarb Anatek-EP 17 Bishop St Portland, ME 04103 email: anatekep@maine.rr.com ------------------------------ Date: Thu, 28 Oct 1999 14:10:23 +0100 (BST) From: ort056@abdn.ac.uk To: nih-image@io.ece.drexel.edu Subject: award Message-ID: Content-Type: TEXT/PLAIN; CHARSET=US-ASCII I agree, There can't be many subscribers who haven't been helped by the people nominated, it would be great to have a beeter way to thank them. ---------------------- Jennifer Gregory j.gregory@abdn.ac.uk Department of Orthopaedics Polwarth Building Foresterhill Aberdeen AB25 2QJ ------------------------------ Date: Thu, 28 Oct 1999 14:41:27 -0600 From: "Berger, Jennifer" To: "'NIH Image List'" Subject: Macro Help Message-ID: Content-Type: text/plain; charset="windows-1252" I am wondering if it is possible (and how to go about doing so) to write a Macro for NIH Image. I have a Scion CG-7 Framegrabber. The problem that I am having is that I need to get my image into 8-Bit color before I can copy it to another program. As things work now I capture in 24-Bit Color, then go to RGB, then finally to 8-Bit. Is there a way to make this one step. Thanks in advance for any help Jennifer ------------------------------ Date: Fri, 29 Oct 1999 09:12:46 +1000 From: Steve Martin To: nih-image@io.ece.drexel.edu Subject: Re: award Message-Id: Content-Type: text/plain; charset="us-ascii" ; format="flowed" At 2:10 PM +0100 on 28/10/99, ort056@abdn.ac.uk wrote: > I agree, There can't be many subscribers who haven't been helped by > the people nominated, it would be great to have a beeter way to thank > them. perhaps it should be called the 'Stained Cup', in reference to all the histological users of NIH Image, or the 'Particular Trophy'. I would be willing to donate a small 3d virtual trophy and a java applet to display it. The annual award holder would be the only person licensed to display it on their web page. I nominate Mr Joss if the annual volume of macro code/pseudocode posted is a criterion. Steve ------------------------------ Date: Fri, 29 Oct 1999 12:13:09 +1030 From: Doug Mason To: nih-image@io.ece.drexel.edu Subject: Re: award Message-Id: Content-Type: text/plain; charset="us-ascii" Steve Martin said: >I nominate Mr Joss if the annual volume of macro code/pseudocode >posted is a criterion. > I second the nomination. I'm sure many others would, too. To the 'annual volume' criterion we could add many other characteristics of Greg Joss's contributions (like insight into problem solving, facility with the macro language, speed of response, etc.). Thanks, Greg! Doug Mason Mason Geoscience Pty Ltd Petrological Services for the Exploration and Mining Industry email : drmason@interconnect.com.au post : PO Box 78, Glenside SA 5065, Australia phone : +61-8-83901507 fax : +61-8-83901194 ------------------------------ Date: Fri, 29 Oct 1999 14:37:29 +1000 From: Jacques Soddell To: nih-image@io.ece.drexel.edu Subject: 3d reconstruction Message-ID: <38192489.275589A7@bendigo.latrobe.edu.au> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit A colleague wishes to make slices through a frog's leg, examine them with a microscope, then capture the images (using a Scion board and NIH-Image) and reconstruct the images into a 3D model. Is there any software or Image macro out there that will help? Thanks -- jacques soddell biological sciences, school of management technology and environment la trobe university bendigo, po box 199, bendigo, victoria, australia. j.soddell@bendigo.latrobe.edu.au http://redgum.bendigo.latrobe.edu.au/~soddell possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 "everything is music, everything is noise" (john cage) http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html -------------------------------- End of nih-image-d Digest V99 Issue #247 **************************************** From nih-image-request@io.ece.drexel.edu Fri Oct 29 00:53 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id AAA26026; Fri, 29 Oct 1999 00:53:41 -0400 (EDT) Resent-Date: Fri, 29 Oct 1999 00:53:41 -0400 (EDT) Message-ID: <38192489.275589A7@bendigo.latrobe.edu.au> Date: Fri, 29 Oct 1999 14:37:29 +1000 From: Jacques Soddell Organization: LaTrobe University, Bendigo X-Mailer: Mozilla 4.61 [en] (Win95; I) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: 3d reconstruction References: <199910280925.FAA10380@io.ece.drexel.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"cuNDR.0.eQ5.OHI6u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1882 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 670 Status: O A colleague wishes to make slices through a frog's leg, examine them with a microscope, then capture the images (using a Scion board and NIH-Image) and reconstruct the images into a 3D model. Is there any software or Image macro out there that will help? Thanks -- jacques soddell biological sciences, school of management technology and environment la trobe university bendigo, po box 199, bendigo, victoria, australia. j.soddell@bendigo.latrobe.edu.au http://redgum.bendigo.latrobe.edu.au/~soddell possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 "everything is music, everything is noise" (john cage) http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html From nih-image-request@io.ece.drexel.edu Fri Oct 29 01:16 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id BAA00778; Fri, 29 Oct 1999 01:16:17 -0400 (EDT) Resent-Date: Fri, 29 Oct 1999 01:16:17 -0400 (EDT) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: J.Soddell@bendigo.latrobe.edu.au, nih-image@io.ece.drexel.edu Date: Fri, 29 Oct 1999 15:07:01 +1000 Subject: Re: 3d reconstruction Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <2EB4B8F16F8@rna.bio.mq.edu.au> Resent-Message-ID: <"XVLKX1.0.Go6.ReI6u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1883 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 1408 Status: O >Date: Fri, 29 Oct 1999 14:37:29 +1000 >From: Jacques Soddell >Organization: LaTrobe University, Bendigo >>To: nih-image@io.ece.drexel.edu >Subject: 3d reconstruction >> >A colleague wishes to make slices through a frog's leg, examine them >with a >microscope, then capture the images (using a Scion board and NIH-Image) >and >reconstruct the images into a 3D model. Is there any software or Image >macro out there that will help? >Thanks > >-- >jacques soddell >biological sciences, school of management technology and environment >la trobe university bendigo, po box 199, bendigo, victoria, australia. > See Object-Image in combination with Rotater available at same location. View tutorials and 3D slicer. Use Roi objects for section outlines. Come back if you have particular questions. Object-Image ------------ by Norbert Vischer, University of Amsterdam e-mail: vischer@bio.uva.nl www: http://simon.bio.uva.nl/object-image.html ftp: ftp://simon.bio.uva.nl/pub/ This program is based on Wayne Rasband's NIH Image --------- www: http://rsb.info.nih.gov/nih-image ftp: ftp://codon.nih.gov/pub/nih-image/ mailing list: nih-image@io.ece.drexel.edu Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Fri Oct 29 12:59 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA15205; Fri, 29 Oct 1999 12:59:18 -0400 (EDT) Resent-Date: Fri, 29 Oct 1999 12:59:18 -0400 (EDT) Message-ID: <3819CE26.8E842C33@welch.jhu.edu> Date: Fri, 29 Oct 1999 12:41:11 -0400 From: "Paul S. Hees" Reply-To: phees@welch.jhu.edu Organization: Johns Hopkins University X-Mailer: Mozilla 4.51 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #247 References: <199910290448.AAA24869@io.ece.drexel.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"6TE__.0.Nk2.srS6u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1884 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Length: 280 Status: O Nested recursive dittos! steve wrote: > I nominate Mr Joss if the annual volume of macro code/pseudocode > posted is a criterion. > > Steve -- Paul S. Hees, Ph.D. MRI Physicist Cardiology Division Johns Hopkins Bayview Medical Center 4940 Eastern Avenue Baltimore, MD 21224 From nih-image-request@io.ece.drexel.edu Fri Oct 29 13:52 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA25560; Fri, 29 Oct 1999 13:52:57 -0400 (EDT) Resent-Date: Fri, 29 Oct 1999 13:52:57 -0400 (EDT) Date: Fri, 29 Oct 1999 10:33:49 -0700 (PDT) From: Lesley Weston To: NIH Image list Subject: Macro to detect blue staining Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"CWB0L3.0.9N5.2gT6u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1885 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 914 Status: O Dear all, We have some cultures of osteoblasts stained for alkaline phosphatase with Fast Blue. We need to measure the blue areas (actually more a kind of purple in various shades). I found Paul C. Grimm's macro in the archives (thanks) and modified it, and it seems to work, but we're not sure if we are measuring all the blue/purple areas and only these areas. If anybody has the time, perhaps you could take a look at it and give us your opinion. This is our version of the macro: macro 'Alkphos stain area [s]'; {sets a number of the variables that need to be set for the image analysis to work properly} begin setoptions('area'); setpalette('system palette'); setscale(0,'pixel'); {changes all the blues in the image to color 107} changevalues(82,123,107); updateLut; {this does the measuring} SetDensitySlice(107,107); wait(.1); measure; {disposeall;} showresults; end; Many thanks. Lesley Weston. From nih-image-request@io.ece.drexel.edu Fri Oct 29 15:08 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id PAA09948; Fri, 29 Oct 1999 15:08:10 -0400 (EDT) Resent-Date: Fri, 29 Oct 1999 15:08:10 -0400 (EDT) Message-ID: <3819EC82.F87EAE41@ucsd.edu> Date: Fri, 29 Oct 1999 11:50:43 -0700 From: Paul Grimm Reply-To: pgrimm@ucsd.edu Organization: UCSD Pediatric Nephrology X-Mailer: Mozilla 4.5 (Macintosh; U; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: Macro to detect blue staining References: Content-Transfer-Encoding: 7bit Resent-Message-ID: <"sxpoW.0.QR1.9oU6u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1886 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Length: 1450 Status: O Hi Lesley Thank you for mentioning my name on the list when you tried using my macro...Can I add this to my CV somehow?...;) Anyway, If you want to see whether you are capturing all the blue areas in the image you can do the following" change the command < wait(.1);> to wait(10); This will give you 10 seconds to inspect which area the macro has selected. (or pick the number of seconds that you feel comfortable with). If you still see blue areas that you believe should be included in the analysis you may need to expand the values that you include in the change values command. for example... changevalues(130,130,35); changevalues(136,136,35); changevalues(141,143,35); changevalues(172,173,35); changevalues(179,179,35); is a series of commands to capture a number of different shades of red in one of my macros. To determine which value you need to include to capture all the colours that you want in the image; use the magnify tool to enlarge the image and read the value of the pixels of interest in the info window when you are developing the color changing macro. Good luck. -- Paul C. Grimm Associate Professor Pediatric Nephrology University of California at San Diego Email pgrimm@ucsd.edu Phone 619-543-5218 Fax 619-543-3575 Snail mail UCSD Pediatrics Mail Stop 0831 9500 Gilman Drive, La Jolla California 92093-0831 Rome did not create a great empire by having meetings... they did it by killing all those who opposed them. From nih-image-d-request@io.ece.drexel.edu Sat Oct 30 06:13 EDT 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA15831; Sat, 30 Oct 1999 06:13:03 -0400 (EDT) Date: Sat, 30 Oct 1999 06:13:03 -0400 (EDT) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199910301013.GAA15831@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #248 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/248 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5875 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 248 Today's Topics: Re: 3d reconstruction [ GJOSS@rna.bio.mq.edu.au ] Re: nih-image-d Digest V99 #247 [ "Paul S. Hees" ] ------------------------------ Date: Fri, 29 Oct 1999 15:07:01 +1000 From: GJOSS@rna.bio.mq.edu.au To: J.Soddell@bendigo.latrobe.edu.au, nih-image@io.ece.drexel.edu Subject: Re: 3d reconstruction Message-ID: <2EB4B8F16F8@rna.bio.mq.edu.au> >Date: Fri, 29 Oct 1999 14:37:29 +1000 >From: Jacques Soddell >Organization: LaTrobe University, Bendigo >>To: nih-image@io.ece.drexel.edu >Subject: 3d reconstruction >> >A colleague wishes to make slices through a frog's leg, examine them >with a >microscope, then capture the images (using a Scion board and NIH-Image) >and >reconstruct the images into a 3D model. Is there any software or Image >macro out there that will help? >Thanks > >-- >jacques soddell >biological sciences, school of management technology and environment >la trobe university bendigo, po box 199, bendigo, victoria, australia. > See Object-Image in combination with Rotater available at same location. View tutorials and 3D slicer. Use Roi objects for section outlines. Come back if you have particular questions. Object-Image ------------ by Norbert Vischer, University of Amsterdam e-mail: vischer@bio.uva.nl www: http://simon.bio.uva.nl/object-image.html ftp: ftp://simon.bio.uva.nl/pub/ This program is based on Wayne Rasband's NIH Image --------- www: http://rsb.info.nih.gov/nih-image ftp: ftp://codon.nih.gov/pub/nih-image/ mailing list: nih-image@io.ece.drexel.edu Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Fri, 29 Oct 1999 12:41:11 -0400 From: "Paul S. Hees" To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #247 Message-ID: <3819CE26.8E842C33@welch.jhu.edu> Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit Nested recursive dittos! steve wrote: > I nominate Mr Joss if the annual volume of macro code/pseudocode > posted is a criterion. > > Steve -- Paul S. Hees, Ph.D. MRI Physicist Cardiology Division Johns Hopkins Bayview Medical Center 4940 Eastern Avenue Baltimore, MD 21224 ------------------------------ Date: Fri, 29 Oct 1999 10:33:49 -0700 (PDT) From: Lesley Weston To: NIH Image list Subject: Macro to detect blue staining Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII Dear all, We have some cultures of osteoblasts stained for alkaline phosphatase with Fast Blue. We need to measure the blue areas (actually more a kind of purple in various shades). I found Paul C. Grimm's macro in the archives (thanks) and modified it, and it seems to work, but we're not sure if we are measuring all the blue/purple areas and only these areas. If anybody has the time, perhaps you could take a look at it and give us your opinion. This is our version of the macro: macro 'Alkphos stain area [s]'; {sets a number of the variables that need to be set for the image analysis to work properly} begin setoptions('area'); setpalette('system palette'); setscale(0,'pixel'); {changes all the blues in the image to color 107} changevalues(82,123,107); updateLut; {this does the measuring} SetDensitySlice(107,107); wait(.1); measure; {disposeall;} showresults; end; Many thanks. Lesley Weston. ------------------------------ Date: Fri, 29 Oct 1999 11:50:43 -0700 From: Paul Grimm To: nih-image@io.ece.drexel.edu Subject: Re: Macro to detect blue staining Message-ID: <3819EC82.F87EAE41@ucsd.edu> Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit Hi Lesley Thank you for mentioning my name on the list when you tried using my macro...Can I add this to my CV somehow?...;) Anyway, If you want to see whether you are capturing all the blue areas in the image you can do the following" change the command < wait(.1);> to wait(10); This will give you 10 seconds to inspect which area the macro has selected. (or pick the number of seconds that you feel comfortable with). If you still see blue areas that you believe should be included in the analysis you may need to expand the values that you include in the change values command. for example... changevalues(130,130,35); changevalues(136,136,35); changevalues(141,143,35); changevalues(172,173,35); changevalues(179,179,35); is a series of commands to capture a number of different shades of red in one of my macros. To determine which value you need to include to capture all the colours that you want in the image; use the magnify tool to enlarge the image and read the value of the pixels of interest in the info window when you are developing the color changing macro. Good luck. -- Paul C. Grimm Associate Professor Pediatric Nephrology University of California at San Diego Email pgrimm@ucsd.edu Phone 619-543-5218 Fax 619-543-3575 Snail mail UCSD Pediatrics Mail Stop 0831 9500 Gilman Drive, La Jolla California 92093-0831 Rome did not create a great empire by having meetings... they did it by killing all those who opposed them. -------------------------------- End of nih-image-d Digest V99 Issue #248 **************************************** From nih-image-request@io.ece.drexel.edu Sun Oct 31 08:16 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA15407; Sun, 31 Oct 1999 08:16:17 -0500 (EST) Resent-Date: Sun, 31 Oct 1999 08:16:17 -0500 (EST) Message-ID: <19991031130013.74878.qmail@hotmail.com> X-Originating-IP: [152.202.96.113] From: "Stephen Motew" To: nih-image@io.ece.drexel.edu Subject: Distance between lines Date: Sun, 31 Oct 1999 05:00:13 PST Mime-Version: 1.0 Resent-Message-ID: <"y6ikb3.0.M53.0s37u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1887 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; format=flowed Content-Length: 552 Status: O Hi! I am trying to measure the mean, max. and min. distances between two straight but not necessarily parallel lines. I trace opposing walls of an arterial wall (from angiogram) and need to define these characterictics for a measured segment. Any help with commands or macros would be greatly appreciated. Thanks. Stephen J. Motew, MD Vascular Surgery Fellow Wake Forest University/ Bowman Gray Medical Center smotew@hotmail.com ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Sun Oct 31 10:58 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id KAA12851; Sun, 31 Oct 1999 10:58:00 -0500 (EST) Resent-Date: Sun, 31 Oct 1999 10:58:00 -0500 (EST) Message-ID: <7D25CCC35D3CD31192BF006008BFB1FE6684EB@nsofs14.uchc.edu> From: "Mansoor,George" To: "'nih-image@io.ece.drexel.edu'" Subject: RE: Distance between lines Date: Sun, 31 Oct 1999 10:42:55 -0500 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Resent-Message-ID: <"1-jcH2.0.qO2.kC67u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1888 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain Content-Length: 990 Status: O Stephen, Give me a call sometime. I think I can help you. George MAnsoor Assistant Professor of Medicine University of Connecticut Health Center 263 Farmington Avenue Farmington, CT. 06030 phone 860-679-2104 > -----Original Message----- > From: Stephen Motew [SMTP:sjmo2@hotmail.com] > Sent: Sunday, October 31, 1999 8:00 AM > To: nih-image@io.ece.drexel.edu > Subject: Distance between lines > > Hi! > > I am trying to measure the mean, max. and min. distances between two > straight but not necessarily parallel lines. I trace opposing walls of an > > arterial wall (from angiogram) and need to define these characterictics > for > a measured segment. Any help with commands or macros would be greatly > appreciated. Thanks. > > > > Stephen J. Motew, MD > Vascular Surgery Fellow > Wake Forest University/ > Bowman Gray Medical Center > smotew@hotmail.com > > ______________________________________________________ > Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Sun Oct 31 18:25 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA22732; Sun, 31 Oct 1999 18:25:05 -0500 (EST) Resent-Date: Sun, 31 Oct 1999 18:25:05 -0500 (EST) Mime-Version: 1.0 X-Sender: scm@128.250.6.196 Message-Id: In-Reply-To: <19991031130013.74878.qmail@hotmail.com> References: <19991031130013.74878.qmail@hotmail.com> Date: Mon, 1 Nov 1999 10:10:35 +1100 To: nih-image@io.ece.drexel.edu From: Steve Martin Subject: Re: Distance between lines Resent-Message-ID: <"AF-bO3.0.Yy4.ToC7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1889 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" ; format="flowed" Content-Length: 1760 Status: O Stephen, if the section you describe is exactly defined by two coplanar straight lines, it is fairly simple geometry to calculate the parameters you require from the end points of the two lines wallAStart,wallAEnd and wallBStart,wallBEnd These x,y coordinates can be very simply obtained with something like: ShowMessage('click at:\start of wall a'); selectPic(angioImage); SetCursor('cross'); repeat getmouse(X,Y); until button; ...etc. using pythagoras (sqrt of sum of squares of x1 to x2 and y1 to y2) to get the normal distances (wallAStart to wallBStart) and (wallAEnd to wallBEnd) the min distance is the minimum of these, the max is the maximum of these and the mean is the average. If the projected image is not coplanar with the walls (e.g. the vessel travels away from the viewer as well as across the view), it gets more complicated than this, and for accurate measurement you will need more than one projection, with common landmarks to calibrate for the effects of perspective. I hope this is some help. Steve At 5:00 AM -0800 on 31/10/99, Stephen Motew wrote: > Hi! > > I am trying to measure the mean, max. and min. distances between >two straight but not necessarily parallel lines. I trace opposing >walls of an arterial wall (from angiogram) and need to define these >characterictics for a measured segment. Any help with commands or >macros would be greatly appreciated. Thanks. > > > > Stephen J. Motew, MD > Vascular Surgery Fellow > Wake Forest University/ > Bowman Gray Medical Center > smotew@hotmail.com ___________________________ Stephen Martin School of Physiotherapy 200 Berkeley St The University of Melbourne Victoria 3010 AUSTRALIA ph +61 3 9344 4171 fax +61 3 9344 4188 ___________________________ From nih-image-request@io.ece.drexel.edu Sun Oct 31 19:13 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA00462; Sun, 31 Oct 1999 19:13:55 -0500 (EST) Resent-Date: Sun, 31 Oct 1999 19:13:55 -0500 (EST) Message-ID: <19991101000020.61969.qmail@hotmail.com> X-Originating-IP: [171.220.117.131] From: "Stephen Motew" To: nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Date: Sun, 31 Oct 1999 16:00:19 PST Mime-Version: 1.0 Resent-Message-ID: <"YDFZb3.0.vv6.sWD7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1890 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; format=flowed Content-Length: 2274 Status: O Thank-you, the images are indeed coplanar (at least for our analysis purposes). I appreciate your assistance and will enjoy the geometry review! -Stephen- >From: Steve Martin >Reply-To: nih-image@io.ece.drexel.edu >To: nih-image@io.ece.drexel.edu >Subject: Re: Distance between lines >Date: Mon, 1 Nov 1999 10:10:35 +1100 > >Stephen, > >if the section you describe is exactly defined by two coplanar >straight lines, it is fairly simple geometry to calculate the >parameters you require from the end points of the two lines > >wallAStart,wallAEnd and wallBStart,wallBEnd > >These x,y coordinates can be very simply obtained with something like: > >ShowMessage('click at:\start of wall a'); >selectPic(angioImage); >SetCursor('cross'); >repeat > getmouse(X,Y); >until button; >...etc. > >using pythagoras (sqrt of sum of squares of x1 to x2 and y1 to y2) to >get the normal distances (wallAStart to wallBStart) and (wallAEnd to >wallBEnd) the min distance is the minimum of these, the max is the >maximum of these and the mean is the average. > >If the projected image is not coplanar with the walls (e.g. the >vessel travels away from the viewer as well as across the view), it >gets more complicated than this, and for accurate measurement you >will need more than one projection, with common landmarks to >calibrate for the effects of perspective. > >I hope this is some help. > >Steve > >At 5:00 AM -0800 on 31/10/99, Stephen Motew wrote: >>Hi! >> >>I am trying to measure the mean, max. and min. distances between >>two straight but not necessarily parallel lines. I trace opposing >>walls of an arterial wall (from angiogram) and need to define these >>characterictics for a measured segment. Any help with commands or >>macros would be greatly appreciated. Thanks. >> >> >> >>Stephen J. Motew, MD >>Vascular Surgery Fellow >>Wake Forest University/ >>Bowman Gray Medical Center >>smotew@hotmail.com >___________________________ >Stephen Martin >School of Physiotherapy >200 Berkeley St >The University of Melbourne >Victoria 3010 >AUSTRALIA >ph +61 3 9344 4171 >fax +61 3 9344 4188 >___________________________ > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Sun Oct 31 20:09 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id UAA10820; Sun, 31 Oct 1999 20:09:29 -0500 (EST) Resent-Date: Sun, 31 Oct 1999 20:09:29 -0500 (EST) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: sjmo2@hotmail.com, smotew@hotmail.com, nih-image@io.ece.drexel.edu Date: Mon, 1 Nov 1999 11:00:07 +1100 Subject: Re: Distance between lines Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <22AA496BEC@rna.bio.mq.edu.au> Resent-Message-ID: <"sEu6S1.0._m1.EJE7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1891 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 2385 Status: O >From: "Stephen Motew" >To: nih-image@io.ece.drexel.edu >Subject: Distance between lines >Date: Sun, 31 Oct 1999 05:00:13 PST > >I am trying to measure the mean, max. and min. distances between two >straight but not necessarily parallel lines. I trace opposing walls of >an arterial wall (from angiogram) and need to define these >characterictics for a measured segment. Any help with commands or >macros would be greatly appreciated. Thanks. > >Stephen J. Motew, MD >Vascular Surgery Fellow >Wake Forest University/ >Bowman Gray Medical Center >smotew@hotmail.com > Stephen, The macro command: binary('edm'); is invaluable in this type of analysis and boundaries may be "any" shape. see "Distance Map(Binary(Process menu". It allows measurement of distance from each pixel with a binary particle to the edge of the particle any shape by converting a binary image into a map where intensity codes the distance from the edge. Obtain a binary image of the artery; eg join ends of arterial lines; fill;threshold; makeBinary; Then apply the following macro: macro'/0edm'; {Greg Joss 1999-11-01 for "Stephen Motew" Assumes binary objects of an elongated "tube" shape so that skeleton of object is a non-branching curve. mean min/max mode radii in results are in pixels. (scaling could be transfered from getScale) Histogram is histogram of radius of tube at each pixel along midline axis. Provided tube axis is either horiziontal or vertical, then pixelcount of midline axis will approximate length along axis. (Length shown in results is actually perimeter.) Otherwise "usual" 2-sqrt(2) correction for "corner" pixels will need to be made. } var bid,eid:integer;begin bid:=pidNumber; duplicate(windowTitle,'.edm'); eid:=pidNumber; binary('edm'); selectPic(bid); duplicate(windowTitle,'.skel'); skeletonize; selectAll;copy;dispose; selectPic(eid); paste;setOption;doAnd; setThreshold(1); setParticleSize(1,99999); setOptions('mean area min/max mode perimeter'); analyzeParticles('reset');undo; showHistogram; showResults; selectPic(eid); selectAll;copy;{dispose;} selectPic(bid); paste;setOption;doReplace; end Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Sun Oct 31 21:39 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id VAA28132; Sun, 31 Oct 1999 21:39:49 -0500 (EST) Resent-Date: Sun, 31 Oct 1999 21:39:49 -0500 (EST) Message-ID: <19991101022639.87620.qmail@hotmail.com> X-Originating-IP: [152.171.106.160] From: "Stephen Motew" To: nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Date: Sun, 31 Oct 1999 18:26:38 PST Mime-Version: 1.0 Resent-Message-ID: <"tH5Gq2.0.EJ6.1gF7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1892 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; format=flowed Content-Length: 3186 Status: O Thank you very much for your help. While running your macro I receive an error on line 20 stating: Unidentified identifier ('edm') Any thoughts? Also it seems if I just use the plot profile data I can derive the "distance" by calculating the number of pixels from the plot. I am having trouble getting a scaled y-axis thought (it just plots it as pixels). Thanks, Stephen >From: GJOSS@rna.bio.mq.edu.au >Reply-To: nih-image@io.ece.drexel.edu >To: sjmo2@hotmail.com, smotew@hotmail.com, nih-image@io.ece.drexel.edu >Subject: Re: Distance between lines >Date: Mon, 1 Nov 1999 11:00:07 +1100 > > >From: "Stephen Motew" > >To: nih-image@io.ece.drexel.edu > >Subject: Distance between lines > >Date: Sun, 31 Oct 1999 05:00:13 PST > > > >I am trying to measure the mean, max. and min. distances between two > >straight but not necessarily parallel lines. I trace opposing walls of > >an arterial wall (from angiogram) and need to define these > >characterictics for a measured segment. Any help with commands or > >macros would be greatly appreciated. Thanks. > > > >Stephen J. Motew, MD > >Vascular Surgery Fellow > >Wake Forest University/ > >Bowman Gray Medical Center > >smotew@hotmail.com > > >Stephen, > >The macro command: binary('edm'); is invaluable in this type of analysis >and boundaries may be "any" shape. >see "Distance Map(Binary(Process menu". > >It allows measurement of distance from each pixel with a binary particle >to the edge of the particle any shape by converting a binary image into >a map where intensity codes the distance from the edge. > >Obtain a binary image of the artery; >eg join ends of arterial lines; fill;threshold; makeBinary; > >Then apply the following macro: > >macro'/0edm'; >{Greg Joss 1999-11-01 >for "Stephen Motew" >Assumes binary objects of an elongated "tube" shape > so that skeleton of object is a non-branching curve. >mean min/max mode radii in results are in pixels. >(scaling could be transfered from getScale) >Histogram is histogram of radius of tube at each pixel >along midline axis. >Provided tube axis is either horiziontal or vertical, >then pixelcount of midline axis will approximate length along axis. >(Length shown in results is actually perimeter.) >Otherwise "usual" 2-sqrt(2) correction for "corner" pixels >will need to be made. >} >var bid,eid:integer;begin >bid:=pidNumber; >duplicate(windowTitle,'.edm'); >eid:=pidNumber; >binary('edm'); >selectPic(bid); >duplicate(windowTitle,'.skel'); >skeletonize; >selectAll;copy;dispose; >selectPic(eid); >paste;setOption;doAnd; >setThreshold(1); >setParticleSize(1,99999); >setOptions('mean area min/max mode perimeter'); >analyzeParticles('reset');undo; >showHistogram; >showResults; >selectPic(eid); >selectAll;copy;{dispose;} >selectPic(bid); >paste;setOption;doReplace; >end > >Greg Joss, >School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 >Macquarie University, Email gjoss@rna.bio.mq.edu.au >North Ryde, (Sydney,) NSW 2109, Australia > > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com From nih-image-request@io.ece.drexel.edu Sun Oct 31 22:40 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id WAA08878; Sun, 31 Oct 1999 22:40:23 -0500 (EST) Resent-Date: Sun, 31 Oct 1999 22:40:23 -0500 (EST) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: sjmo2@hotmail.com, nih-image@io.ece.drexel.edu Date: Mon, 1 Nov 1999 13:33:19 +1100 Subject: Re: Distance between lines Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <25384D7018@rna.bio.mq.edu.au> Resent-Message-ID: <"4cIta2.0.0b1.vYG7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1893 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 1529 Status: O >From: "Stephen Motew" >To: nih-image@io.ece.drexel.edu >Subject: Re: Distance between lines >Date: Sun, 31 Oct 1999 18:26:38 PST > >Thank you very much for your help. > >While running your macro I receive an error on line 20 stating: > >Unidentified identifier ('edm') > >Any thoughts? I suspect that you must be running a (very) old version of image prior to the binary ('edm'),ultimate points and watershed version. You should download V1.62 or Object-Image ------------ by Norbert Vischer, University of Amsterdam e-mail: vischer@bio.uva.nl www: http://simon.bio.uva.nl/object-image.html ftp: ftp://simon.bio.uva.nl/pub/ This program is based on Wayne Rasband's NIH Image --------- www: http://rsb.info.nih.gov/nih-image ftp: ftp://codon.nih.gov/pub/nih-image/ mailing list: nih-image@io.ece.drexel.edu >Also it seems if I just use the plot profile data I can derive the >"distance" by calculating the number of pixels from the plot. I am >having >trouble getting a scaled y-axis thought (it just plots it as pixels). > This method would a count of pixels at rightangles to the profile selection so that you would be limited to straight lines. You could scale by postprocess of plotdata array values. ie multiply pixel counts by getscale data. >Thanks, > >Stephen Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-d-request@io.ece.drexel.edu Mon Nov 1 06:18 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA04886; Mon, 1 Nov 1999 06:18:50 -0500 (EST) Date: Mon, 1 Nov 1999 06:18:50 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911011118.GAA04886@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #249 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/249 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 15391 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 249 Today's Topics: Distance between lines [ "Stephen Motew" ] RE: Distance between lines [ "Mansoor,George" ] Re: Distance between lines [ GJOSS@rna.bio.mq.edu.au ] Re: Distance between lines [ "Stephen Motew" ] Re: Distance between lines [ GJOSS@rna.bio.mq.edu.au ] ------------------------------ Date: Sun, 31 Oct 1999 05:00:13 PST From: "Stephen Motew" To: nih-image@io.ece.drexel.edu Subject: Distance between lines Message-ID: <19991031130013.74878.qmail@hotmail.com> Content-Type: text/plain; format=flowed Hi! I am trying to measure the mean, max. and min. distances between two straight but not necessarily parallel lines. I trace opposing walls of an arterial wall (from angiogram) and need to define these characterictics for a measured segment. Any help with commands or macros would be greatly appreciated. Thanks. Stephen J. Motew, MD Vascular Surgery Fellow Wake Forest University/ Bowman Gray Medical Center smotew@hotmail.com ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com ------------------------------ Date: Sun, 31 Oct 1999 10:42:55 -0500 From: "Mansoor,George" To: "'nih-image@io.ece.drexel.edu'" Subject: RE: Distance between lines Message-ID: <7D25CCC35D3CD31192BF006008BFB1FE6684EB@nsofs14.uchc.edu> Content-Type: text/plain Stephen, Give me a call sometime. I think I can help you. George MAnsoor Assistant Professor of Medicine University of Connecticut Health Center 263 Farmington Avenue Farmington, CT. 06030 phone 860-679-2104 > -----Original Message----- > From: Stephen Motew [SMTP:sjmo2@hotmail.com] > Sent: Sunday, October 31, 1999 8:00 AM > To: nih-image@io.ece.drexel.edu > Subject: Distance between lines > > Hi! > > I am trying to measure the mean, max. and min. distances between two > straight but not necessarily parallel lines. I trace opposing walls of an > > arterial wall (from angiogram) and need to define these characterictics > for > a measured segment. Any help with commands or macros would be greatly > appreciated. Thanks. > > > > Stephen J. Motew, MD > Vascular Surgery Fellow > Wake Forest University/ > Bowman Gray Medical Center > smotew@hotmail.com > > ______________________________________________________ > Get Your Private, Free Email at http://www.hotmail.com ------------------------------ Date: Mon, 1 Nov 1999 10:10:35 +1100 From: Steve Martin To: nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Message-Id: Content-Type: text/plain; charset="us-ascii" ; format="flowed" Stephen, if the section you describe is exactly defined by two coplanar straight lines, it is fairly simple geometry to calculate the parameters you require from the end points of the two lines wallAStart,wallAEnd and wallBStart,wallBEnd These x,y coordinates can be very simply obtained with something like: ShowMessage('click at:\start of wall a'); selectPic(angioImage); SetCursor('cross'); repeat getmouse(X,Y); until button; ...etc. using pythagoras (sqrt of sum of squares of x1 to x2 and y1 to y2) to get the normal distances (wallAStart to wallBStart) and (wallAEnd to wallBEnd) the min distance is the minimum of these, the max is the maximum of these and the mean is the average. If the projected image is not coplanar with the walls (e.g. the vessel travels away from the viewer as well as across the view), it gets more complicated than this, and for accurate measurement you will need more than one projection, with common landmarks to calibrate for the effects of perspective. I hope this is some help. Steve At 5:00 AM -0800 on 31/10/99, Stephen Motew wrote: > Hi! > > I am trying to measure the mean, max. and min. distances between >two straight but not necessarily parallel lines. I trace opposing >walls of an arterial wall (from angiogram) and need to define these >characterictics for a measured segment. Any help with commands or >macros would be greatly appreciated. Thanks. > > > > Stephen J. Motew, MD > Vascular Surgery Fellow > Wake Forest University/ > Bowman Gray Medical Center > smotew@hotmail.com ___________________________ Stephen Martin School of Physiotherapy 200 Berkeley St The University of Melbourne Victoria 3010 AUSTRALIA ph +61 3 9344 4171 fax +61 3 9344 4188 ___________________________ ------------------------------ Date: Sun, 31 Oct 1999 16:00:19 PST From: "Stephen Motew" To: nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Message-ID: <19991101000020.61969.qmail@hotmail.com> Content-Type: text/plain; format=flowed Thank-you, the images are indeed coplanar (at least for our analysis purposes). I appreciate your assistance and will enjoy the geometry review! -Stephen- >From: Steve Martin >Reply-To: nih-image@io.ece.drexel.edu >To: nih-image@io.ece.drexel.edu >Subject: Re: Distance between lines >Date: Mon, 1 Nov 1999 10:10:35 +1100 > >Stephen, > >if the section you describe is exactly defined by two coplanar >straight lines, it is fairly simple geometry to calculate the >parameters you require from the end points of the two lines > >wallAStart,wallAEnd and wallBStart,wallBEnd > >These x,y coordinates can be very simply obtained with something like: > >ShowMessage('click at:\start of wall a'); >selectPic(angioImage); >SetCursor('cross'); >repeat > getmouse(X,Y); >until button; >...etc. > >using pythagoras (sqrt of sum of squares of x1 to x2 and y1 to y2) to >get the normal distances (wallAStart to wallBStart) and (wallAEnd to >wallBEnd) the min distance is the minimum of these, the max is the >maximum of these and the mean is the average. > >If the projected image is not coplanar with the walls (e.g. the >vessel travels away from the viewer as well as across the view), it >gets more complicated than this, and for accurate measurement you >will need more than one projection, with common landmarks to >calibrate for the effects of perspective. > >I hope this is some help. > >Steve > >At 5:00 AM -0800 on 31/10/99, Stephen Motew wrote: >>Hi! >> >>I am trying to measure the mean, max. and min. distances between >>two straight but not necessarily parallel lines. I trace opposing >>walls of an arterial wall (from angiogram) and need to define these >>characterictics for a measured segment. Any help with commands or >>macros would be greatly appreciated. Thanks. >> >> >> >>Stephen J. Motew, MD >>Vascular Surgery Fellow >>Wake Forest University/ >>Bowman Gray Medical Center >>smotew@hotmail.com >___________________________ >Stephen Martin >School of Physiotherapy >200 Berkeley St >The University of Melbourne >Victoria 3010 >AUSTRALIA >ph +61 3 9344 4171 >fax +61 3 9344 4188 >___________________________ > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com ------------------------------ Date: Mon, 1 Nov 1999 11:00:07 +1100 From: GJOSS@rna.bio.mq.edu.au To: sjmo2@hotmail.com, smotew@hotmail.com, nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Message-ID: <22AA496BEC@rna.bio.mq.edu.au> >From: "Stephen Motew" >To: nih-image@io.ece.drexel.edu >Subject: Distance between lines >Date: Sun, 31 Oct 1999 05:00:13 PST > >I am trying to measure the mean, max. and min. distances between two >straight but not necessarily parallel lines. I trace opposing walls of >an arterial wall (from angiogram) and need to define these >characterictics for a measured segment. Any help with commands or >macros would be greatly appreciated. Thanks. > >Stephen J. Motew, MD >Vascular Surgery Fellow >Wake Forest University/ >Bowman Gray Medical Center >smotew@hotmail.com > Stephen, The macro command: binary('edm'); is invaluable in this type of analysis and boundaries may be "any" shape. see "Distance Map(Binary(Process menu". It allows measurement of distance from each pixel with a binary particle to the edge of the particle any shape by converting a binary image into a map where intensity codes the distance from the edge. Obtain a binary image of the artery; eg join ends of arterial lines; fill;threshold; makeBinary; Then apply the following macro: macro'/0edm'; {Greg Joss 1999-11-01 for "Stephen Motew" Assumes binary objects of an elongated "tube" shape so that skeleton of object is a non-branching curve. mean min/max mode radii in results are in pixels. (scaling could be transfered from getScale) Histogram is histogram of radius of tube at each pixel along midline axis. Provided tube axis is either horiziontal or vertical, then pixelcount of midline axis will approximate length along axis. (Length shown in results is actually perimeter.) Otherwise "usual" 2-sqrt(2) correction for "corner" pixels will need to be made. } var bid,eid:integer;begin bid:=pidNumber; duplicate(windowTitle,'.edm'); eid:=pidNumber; binary('edm'); selectPic(bid); duplicate(windowTitle,'.skel'); skeletonize; selectAll;copy;dispose; selectPic(eid); paste;setOption;doAnd; setThreshold(1); setParticleSize(1,99999); setOptions('mean area min/max mode perimeter'); analyzeParticles('reset');undo; showHistogram; showResults; selectPic(eid); selectAll;copy;{dispose;} selectPic(bid); paste;setOption;doReplace; end Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Sun, 31 Oct 1999 18:26:38 PST From: "Stephen Motew" To: nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Message-ID: <19991101022639.87620.qmail@hotmail.com> Content-Type: text/plain; format=flowed Thank you very much for your help. While running your macro I receive an error on line 20 stating: Unidentified identifier ('edm') Any thoughts? Also it seems if I just use the plot profile data I can derive the "distance" by calculating the number of pixels from the plot. I am having trouble getting a scaled y-axis thought (it just plots it as pixels). Thanks, Stephen >From: GJOSS@rna.bio.mq.edu.au >Reply-To: nih-image@io.ece.drexel.edu >To: sjmo2@hotmail.com, smotew@hotmail.com, nih-image@io.ece.drexel.edu >Subject: Re: Distance between lines >Date: Mon, 1 Nov 1999 11:00:07 +1100 > > >From: "Stephen Motew" > >To: nih-image@io.ece.drexel.edu > >Subject: Distance between lines > >Date: Sun, 31 Oct 1999 05:00:13 PST > > > >I am trying to measure the mean, max. and min. distances between two > >straight but not necessarily parallel lines. I trace opposing walls of > >an arterial wall (from angiogram) and need to define these > >characterictics for a measured segment. Any help with commands or > >macros would be greatly appreciated. Thanks. > > > >Stephen J. Motew, MD > >Vascular Surgery Fellow > >Wake Forest University/ > >Bowman Gray Medical Center > >smotew@hotmail.com > > >Stephen, > >The macro command: binary('edm'); is invaluable in this type of analysis >and boundaries may be "any" shape. >see "Distance Map(Binary(Process menu". > >It allows measurement of distance from each pixel with a binary particle >to the edge of the particle any shape by converting a binary image into >a map where intensity codes the distance from the edge. > >Obtain a binary image of the artery; >eg join ends of arterial lines; fill;threshold; makeBinary; > >Then apply the following macro: > >macro'/0edm'; >{Greg Joss 1999-11-01 >for "Stephen Motew" >Assumes binary objects of an elongated "tube" shape > so that skeleton of object is a non-branching curve. >mean min/max mode radii in results are in pixels. >(scaling could be transfered from getScale) >Histogram is histogram of radius of tube at each pixel >along midline axis. >Provided tube axis is either horiziontal or vertical, >then pixelcount of midline axis will approximate length along axis. >(Length shown in results is actually perimeter.) >Otherwise "usual" 2-sqrt(2) correction for "corner" pixels >will need to be made. >} >var bid,eid:integer;begin >bid:=pidNumber; >duplicate(windowTitle,'.edm'); >eid:=pidNumber; >binary('edm'); >selectPic(bid); >duplicate(windowTitle,'.skel'); >skeletonize; >selectAll;copy;dispose; >selectPic(eid); >paste;setOption;doAnd; >setThreshold(1); >setParticleSize(1,99999); >setOptions('mean area min/max mode perimeter'); >analyzeParticles('reset');undo; >showHistogram; >showResults; >selectPic(eid); >selectAll;copy;{dispose;} >selectPic(bid); >paste;setOption;doReplace; >end > >Greg Joss, >School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 >Macquarie University, Email gjoss@rna.bio.mq.edu.au >North Ryde, (Sydney,) NSW 2109, Australia > > ______________________________________________________ Get Your Private, Free Email at http://www.hotmail.com ------------------------------ Date: Mon, 1 Nov 1999 13:33:19 +1100 From: GJOSS@rna.bio.mq.edu.au To: sjmo2@hotmail.com, nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Message-ID: <25384D7018@rna.bio.mq.edu.au> >From: "Stephen Motew" >To: nih-image@io.ece.drexel.edu >Subject: Re: Distance between lines >Date: Sun, 31 Oct 1999 18:26:38 PST > >Thank you very much for your help. > >While running your macro I receive an error on line 20 stating: > >Unidentified identifier ('edm') > >Any thoughts? I suspect that you must be running a (very) old version of image prior to the binary ('edm'),ultimate points and watershed version. You should download V1.62 or Object-Image ------------ by Norbert Vischer, University of Amsterdam e-mail: vischer@bio.uva.nl www: http://simon.bio.uva.nl/object-image.html ftp: ftp://simon.bio.uva.nl/pub/ This program is based on Wayne Rasband's NIH Image --------- www: http://rsb.info.nih.gov/nih-image ftp: ftp://codon.nih.gov/pub/nih-image/ mailing list: nih-image@io.ece.drexel.edu >Also it seems if I just use the plot profile data I can derive the >"distance" by calculating the number of pixels from the plot. I am >having >trouble getting a scaled y-axis thought (it just plots it as pixels). > This method would a count of pixels at rightangles to the profile selection so that you would be limited to straight lines. You could scale by postprocess of plotdata array values. ie multiply pixel counts by getscale data. >Thanks, > >Stephen Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia -------------------------------- End of nih-image-d Digest V99 Issue #249 **************************************** From nih-image-request@io.ece.drexel.edu Mon Nov 1 11:29 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA09382; Mon, 1 Nov 1999 11:29:49 -0500 (EST) Resent-Date: Mon, 1 Nov 1999 11:29:49 -0500 (EST) X-Authentication-Warning: mail.bio.uva.nl: Host gold.bio.uva.nl [145.18.160.48] claimed to be [145.18.160.48] Message-Id: Mime-Version: 1.0 Date: Mon, 1 Nov 1999 17:07:24 +0100 To: nih-image@io.ece.drexel.edu From: Norbert Vischer Subject: Re: Distance between lines Resent-Message-ID: <"Efnt9.0.547.beR7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1894 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 7316 Status: O Stephen, here is a solution which calculates distances geometrically, is non-destructive and is able to *visualize* the distance lines whose lenghts you want to know. It also calculates the "score" per target when more than one targets exist, and identifies which point "sparks" to which target. As Greg Joss already mentioned, Euclidian Distance Map is the most straight forward method to calculate the closest distance between a point and a (border-) line. However, I had recently to work on a problem where EDM didn't suffice. I have now added this example to my list of documented examples. The examples on my site are all accompanied with a readme file which describes what it does, how to try it out and how the experiment was prepared from scratch. I include here the readme file which is part of ftp://simon.bio.uva.nl/pub/Sparks.sea.hqx Sparks.obj ---------- What it does: ============= Visualizes spark lines and calculates their statistics. A spark is here the shortest possible vector from a point object to one or several segmented line objects, which are here called 'targets'. Also polygons or rois can be used as targets. Rather than using an Euclidian distance map, the sparks are calculated geometrically with sub-pixel resolution and not limited to 255, additionally yielding their direction and impact points. In this example, the closest distance from vesicles to several membranes had to be measured. The membranes were considered as targets. How to try it out: ================== - Open Object-Image1.62p7 or later - In the Finder or from Object-Image, open the file "Sparks.obj". - An "Info card" comes up and indicates that an image called "Vesicles-c" already contains some markers. Double-click this file name. The image (located in the same folder as the object file) becomes visible together with non-destructive markers. The magenta markers are segmented line objects, which are defined as targets. Also a number of point objects in red and blue color are visible. - Issue "Calculate Sparks" from the special menu. For all point objects in all images, the closest distances to any of the targets are now calculated. - The data are now up-to-date, and you can visualize the spark lines: From the special menu, choose "Draw Spark Lines". This causes the spark lines to appear in red in the front-most image. (They are mapped to the LSB of the image). - Try "Toggle LUT" from the Special menu to show/hide the spark lines. This command sets odd Lut entries either to red or resets them to normal greyscale. - Choose "Distance Histogram" to show a histogram of the lengths of all spark lines. Then close this graph without saving. - The program is designed to perform measurements across many images and/or stack slices. To set more markers, bring "Vesicles-c" or another image to the front : - Choose 'Toggle LUT' from the Special menu until the spark lines disappear. - Click on the "Obj" tool in the Tools window (bottom). - The floating "Sequence" window becomes visible, showing the object types which are currently available (like "mebrane" and "vesicle"). You can click the sequence window's top right corner to paritally hide or fully show it. Click e.g on the "Vesicle" object in the Sequence window (or press key "v"), and set some more points into the image. Similarly, you can choose "Membrane" and add more targets. Use the tab or return key to finish a segmented line. - Remove any object by shooting it away with the pistol tool (bottom of extended Tools window). - To calculate the results, choose "Calculate Sparks". - Visualize the results by choosing "Show Obj. Results": (use option key to include statistics) The columns cointain these data: Distance: spark distances in pixels SDistance: spark distances in microns ImpactX, ImpactY: impact points of a spark on a target Score: numbers of sparks arriving at a target Target: the IDs of the found targets - Choose "Save Objects" from the Objects menu, or click on the "Save Obj" button in the Sequence window to save markers and results in the current object file (here spark.obj). - The images are restricted to even values 8.. 252 and loose their odd values due to these operations. This doesn't matter in many cases, but you can avoid to save the changed images without any penalty and so remain completely non-destructive. Take care not to save an image as PICT while the red spark lines are visible, because PICT saves colors, not pixel values. How this experiment was prepared: ================================= - In Object-Image1.62p7, I chose "New Object file" from the objects menu and named it "Sparks.obj". In the dialog, I left the "keep.." checkboxes untouched. Then I closed this dialog and came automatically into the "Define Objects" dialog by agreeing "Define object now?" - In the "Define Objects" Dialog, I dragged a "SegLine" from "Object types" into the "Sequence" area. Then I dragged 4 times a point object type into the "Sequence area". _ I clicked the "Single" radio button, because I wanted to carry each object its own "cell" number. "3d objects" remained unchecked. - Then I began to modify the 5 items in the Sequence area: I double-clicked the first item and changed the name from "SegLineA" to "Membrane". I clicked the appearance buttons to change color and line type. I clicked the "R" button" and chose "No Result". - Similarly, I renamed the point objects, changed appearance and chose "No Result" because I was not interested in any of the automatic results. A result column can also be removed by selecting it an pressing the backspace key. Finally, the Results area was empty. - I added two keyboard shortcuts by clicking the "Key" button on two items, and chose letters which were not occupied by macro shortcuts. - Rather than adding result columns with the "New Result" button, I left it up to the macro to create the desired columns (using 'InitColumn'); - Now I closed the Dialog with OK, and saved the object file via Save Objects from the Objects menu. I located the object file into a dedicated folder, which also contained the images to be marked. - Finally, I opened the text card via menu "Objects - Text/Macro Card" and typed in the macros. The macros are set to be autoloaded, i.e. they appear automatically under the Special menu when the object file is loaded from the disk. Because the target is of type "segmented line" and not "Roi" or "Polygon", I set the boolean "ClosedTarget" in the macro text to "false". To study the macros, you can option-double click any of the keywords to get an on-line explanation, or you can issue a macro from the menu with the command key down, to observe image, macro text and variables simultaneously, while single-stepping through the macro. Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html From nih-image-request@io.ece.drexel.edu Mon Nov 1 13:30 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA01630; Mon, 1 Nov 1999 13:30:51 -0500 (EST) Resent-Date: Mon, 1 Nov 1999 13:30:51 -0500 (EST) Date: Mon, 1 Nov 1999 23:44:34 +0530 (IST) From: Ajai Vyas To: NIHlist Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"jji5A.0.Tb6.cST7u"@io> Resent-From: nih-image@io.ece.drexel.edu Subject: Unidentified subject! Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1895 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 1116 Status: O Hi, I am new to image measurements and just wanted to confirm I am doing right things before I interpret my results. I want to measure total dendritic lengths of neurons from scanned traces of them, originally drawn by camera lucida on a paper along with a calibration bar of 100 microns. I measure the number of pixels in calibration line (n), then skeletonize image of dendrite, count number of pixels and divide by n to get length. When I select freehand tool and trace the length on image, number obtained by this seems to be higher than that in earlier method. What reason it might have or there is anything fundamentally wrong in my approach? Thanks for help, AJAI. ------------------------------------------------------------------------------- AJAI VYAS, National Center For Biological Sciences, Phone-080-8560459 UAS-GKVK Campus, 080-8561654 Bangalore-560065, INDIA. Ext. -3221 & 5009 E-mail - ajai@ncbs.res.in Fax - 080-8561662 - ajaivyas@hotmail.com ------------------------------------------------------------------------------- From nih-image-request@io.ece.drexel.edu Mon Nov 1 15:56 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id PAA26730; Mon, 1 Nov 1999 15:56:44 -0500 (EST) Resent-Date: Mon, 1 Nov 1999 15:56:44 -0500 (EST) Message-Id: <199911012026.PAA01111@oeb.harvard.edu> X-Sender: spicer@pop.oeb.harvard.edu X-Mailer: QUALCOMM Windows Eudora Pro Version 4.0.2 Date: Mon, 01 Nov 1999 15:36:47 -0500 To: nih-image@io.ece.drexel.edu From: Rachel Spicer Subject: Scion Image experience? Mime-Version: 1.0 Resent-Message-ID: <"scg1s1.0.Jo5.wdV7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1896 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 708 Status: O Hi all- I've just left a MAC-based lab that used NIH Image and hope to set up a similar imaging facility (currently PC-based) at my new institution. I'm wondering if anyone out there has experience with the Scion Image program available for the PC. I remember the beta versions were full of bugs, and I am curious how the current version compares to NIH Image v. 1.61. Any thoughts? Thanks in advance, Rachel ****************************************** Rachel Spicer Biological Laboratories 395 Organismic and Evolutionary Biology Harvard University 16 Divinity Avenue Cambridge, MA 02138 (617) 496-3580 (phone) (617) 496-5854 (fax) spicer@oeb.harvard.edu ****************************************** From nih-image-request@io.ece.drexel.edu Mon Nov 1 16:34 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA05699; Mon, 1 Nov 1999 16:34:27 -0500 (EST) Resent-Date: Mon, 1 Nov 1999 16:34:27 -0500 (EST) Message-Id: In-Reply-To: Mime-Version: 1.0 Date: Mon, 1 Nov 1999 04:20:56 -0500 To: nih-image@io.ece.drexel.edu From: Arnout Ruifrok Subject: Re: Unidentified subject! Resent-Message-ID: <"JU3GN3.0.eR.cCW7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1897 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1280 Status: O >Hi, > I am new to image measurements and just wanted to confirm I am >doing right things before I interpret my results. I want to measure total >dendritic lengths of neurons from scanned traces of them, originally drawn >by camera lucida on a paper along with a calibration bar of 100 microns. I >measure the number of pixels in calibration line (n), then skeletonize >image of dendrite, count number of pixels and divide by n to get length. > When I select freehand tool and trace the length on image, number >obtained by this seems to be higher than that in earlier method. What >reason it might have or there is anything fundamentally wrong in my >approach? > Thanks for help, >AJAI. >------------------------------------------------------------------------------- > >AJAI VYAS, >National Center For Biological Sciences, Phone-080-8560459 >UAS-GKVK Campus, 080-8561654 >Bangalore-560065, INDIA. Ext. -3221 & 5009 >E-mail - ajai@ncbs.res.in Fax - 080-8561662 > - ajaivyas@hotmail.com > >------------------------------------------------------------------------------- The freehand measurements corrects for diagonal connected pixels (sqrt 2), your pixels count does not correct for diagonal; connection and will under estimate the lenght. Arnout From nih-image-d-request@io.ece.drexel.edu Tue Nov 2 05:31 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA23264; Tue, 2 Nov 1999 05:31:05 -0500 (EST) Date: Tue, 2 Nov 1999 05:31:05 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911021031.FAA23264@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #250 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/250 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 13611 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 250 Today's Topics: Re: Distance between lines [ Norbert Vischer ] Scion Image experience? [ Rachel Spicer ] ------------------------------ Date: Mon, 1 Nov 1999 17:07:24 +0100 From: Norbert Vischer To: nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Message-Id: Content-Type: text/plain; charset="us-ascii" Stephen, here is a solution which calculates distances geometrically, is non-destructive and is able to *visualize* the distance lines whose lenghts you want to know. It also calculates the "score" per target when more than one targets exist, and identifies which point "sparks" to which target. As Greg Joss already mentioned, Euclidian Distance Map is the most straight forward method to calculate the closest distance between a point and a (border-) line. However, I had recently to work on a problem where EDM didn't suffice. I have now added this example to my list of documented examples. The examples on my site are all accompanied with a readme file which describes what it does, how to try it out and how the experiment was prepared from scratch. I include here the readme file which is part of ftp://simon.bio.uva.nl/pub/Sparks.sea.hqx Sparks.obj ---------- What it does: ============= Visualizes spark lines and calculates their statistics. A spark is here the shortest possible vector from a point object to one or several segmented line objects, which are here called 'targets'. Also polygons or rois can be used as targets. Rather than using an Euclidian distance map, the sparks are calculated geometrically with sub-pixel resolution and not limited to 255, additionally yielding their direction and impact points. In this example, the closest distance from vesicles to several membranes had to be measured. The membranes were considered as targets. How to try it out: ================== - Open Object-Image1.62p7 or later - In the Finder or from Object-Image, open the file "Sparks.obj". - An "Info card" comes up and indicates that an image called "Vesicles-c" already contains some markers. Double-click this file name. The image (located in the same folder as the object file) becomes visible together with non-destructive markers. The magenta markers are segmented line objects, which are defined as targets. Also a number of point objects in red and blue color are visible. - Issue "Calculate Sparks" from the special menu. For all point objects in all images, the closest distances to any of the targets are now calculated. - The data are now up-to-date, and you can visualize the spark lines: From the special menu, choose "Draw Spark Lines". This causes the spark lines to appear in red in the front-most image. (They are mapped to the LSB of the image). - Try "Toggle LUT" from the Special menu to show/hide the spark lines. This command sets odd Lut entries either to red or resets them to normal greyscale. - Choose "Distance Histogram" to show a histogram of the lengths of all spark lines. Then close this graph without saving. - The program is designed to perform measurements across many images and/or stack slices. To set more markers, bring "Vesicles-c" or another image to the front : - Choose 'Toggle LUT' from the Special menu until the spark lines disappear. - Click on the "Obj" tool in the Tools window (bottom). - The floating "Sequence" window becomes visible, showing the object types which are currently available (like "mebrane" and "vesicle"). You can click the sequence window's top right corner to paritally hide or fully show it. Click e.g on the "Vesicle" object in the Sequence window (or press key "v"), and set some more points into the image. Similarly, you can choose "Membrane" and add more targets. Use the tab or return key to finish a segmented line. - Remove any object by shooting it away with the pistol tool (bottom of extended Tools window). - To calculate the results, choose "Calculate Sparks". - Visualize the results by choosing "Show Obj. Results": (use option key to include statistics) The columns cointain these data: Distance: spark distances in pixels SDistance: spark distances in microns ImpactX, ImpactY: impact points of a spark on a target Score: numbers of sparks arriving at a target Target: the IDs of the found targets - Choose "Save Objects" from the Objects menu, or click on the "Save Obj" button in the Sequence window to save markers and results in the current object file (here spark.obj). - The images are restricted to even values 8.. 252 and loose their odd values due to these operations. This doesn't matter in many cases, but you can avoid to save the changed images without any penalty and so remain completely non-destructive. Take care not to save an image as PICT while the red spark lines are visible, because PICT saves colors, not pixel values. How this experiment was prepared: ================================= - In Object-Image1.62p7, I chose "New Object file" from the objects menu and named it "Sparks.obj". In the dialog, I left the "keep.." checkboxes untouched. Then I closed this dialog and came automatically into the "Define Objects" dialog by agreeing "Define object now?" - In the "Define Objects" Dialog, I dragged a "SegLine" from "Object types" into the "Sequence" area. Then I dragged 4 times a point object type into the "Sequence area". _ I clicked the "Single" radio button, because I wanted to carry each object its own "cell" number. "3d objects" remained unchecked. - Then I began to modify the 5 items in the Sequence area: I double-clicked the first item and changed the name from "SegLineA" to "Membrane". I clicked the appearance buttons to change color and line type. I clicked the "R" button" and chose "No Result". - Similarly, I renamed the point objects, changed appearance and chose "No Result" because I was not interested in any of the automatic results. A result column can also be removed by selecting it an pressing the backspace key. Finally, the Results area was empty. - I added two keyboard shortcuts by clicking the "Key" button on two items, and chose letters which were not occupied by macro shortcuts. - Rather than adding result columns with the "New Result" button, I left it up to the macro to create the desired columns (using 'InitColumn'); - Now I closed the Dialog with OK, and saved the object file via Save Objects from the Objects menu. I located the object file into a dedicated folder, which also contained the images to be marked. - Finally, I opened the text card via menu "Objects - Text/Macro Card" and typed in the macros. The macros are set to be autoloaded, i.e. they appear automatically under the Special menu when the object file is loaded from the disk. Because the target is of type "segmented line" and not "Roi" or "Polygon", I set the boolean "ClosedTarget" in the macro text to "false". To study the macros, you can option-double click any of the keywords to get an on-line explanation, or you can issue a macro from the menu with the command key down, to observe image, macro text and variables simultaneously, while single-stepping through the macro. Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html ------------------------------ Date: Mon, 1 Nov 1999 23:44:34 +0530 (IST) From: Ajai Vyas To: NIHlist Subject: Unidentified subject! Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I am new to image measurements and just wanted to confirm I am doing right things before I interpret my results. I want to measure total dendritic lengths of neurons from scanned traces of them, originally drawn by camera lucida on a paper along with a calibration bar of 100 microns. I measure the number of pixels in calibration line (n), then skeletonize image of dendrite, count number of pixels and divide by n to get length. When I select freehand tool and trace the length on image, number obtained by this seems to be higher than that in earlier method. What reason it might have or there is anything fundamentally wrong in my approach? Thanks for help, AJAI. ------------------------------------------------------------------------------- AJAI VYAS, National Center For Biological Sciences, Phone-080-8560459 UAS-GKVK Campus, 080-8561654 Bangalore-560065, INDIA. Ext. -3221 & 5009 E-mail - ajai@ncbs.res.in Fax - 080-8561662 - ajaivyas@hotmail.com ------------------------------------------------------------------------------- ------------------------------ Date: Mon, 01 Nov 1999 15:36:47 -0500 From: Rachel Spicer To: nih-image@io.ece.drexel.edu Subject: Scion Image experience? Message-Id: <199911012026.PAA01111@oeb.harvard.edu> Content-Type: text/plain; charset="us-ascii" Hi all- I've just left a MAC-based lab that used NIH Image and hope to set up a similar imaging facility (currently PC-based) at my new institution. I'm wondering if anyone out there has experience with the Scion Image program available for the PC. I remember the beta versions were full of bugs, and I am curious how the current version compares to NIH Image v. 1.61. Any thoughts? Thanks in advance, Rachel ****************************************** Rachel Spicer Biological Laboratories 395 Organismic and Evolutionary Biology Harvard University 16 Divinity Avenue Cambridge, MA 02138 (617) 496-3580 (phone) (617) 496-5854 (fax) spicer@oeb.harvard.edu ****************************************** ------------------------------ Date: Mon, 1 Nov 1999 04:20:56 -0500 From: Arnout Ruifrok To: nih-image@io.ece.drexel.edu Subject: Re: Unidentified subject! Message-Id: Content-Type: text/plain; charset="us-ascii" >Hi, > I am new to image measurements and just wanted to confirm I am >doing right things before I interpret my results. I want to measure total >dendritic lengths of neurons from scanned traces of them, originally drawn >by camera lucida on a paper along with a calibration bar of 100 microns. I >measure the number of pixels in calibration line (n), then skeletonize >image of dendrite, count number of pixels and divide by n to get length. > When I select freehand tool and trace the length on image, number >obtained by this seems to be higher than that in earlier method. What >reason it might have or there is anything fundamentally wrong in my >approach? > Thanks for help, >AJAI. >------------------------------------------------------------------------------- > >AJAI VYAS, >National Center For Biological Sciences, Phone-080-8560459 >UAS-GKVK Campus, 080-8561654 >Bangalore-560065, INDIA. Ext. -3221 & 5009 >E-mail - ajai@ncbs.res.in Fax - 080-8561662 > - ajaivyas@hotmail.com > >------------------------------------------------------------------------------- The freehand measurements corrects for diagonal connected pixels (sqrt 2), your pixels count does not correct for diagonal; connection and will under estimate the lenght. Arnout ------------------------------ Date: Tue, 2 Nov 1999 11:18:35 +0100 From: Gary Chinga To: "'nih-image@io.ece.drexel.edu'" Subject: RE: 3d reconstruction Message-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit You can use NIH Image or better Object Image to make reconstructions, take a look at this page and see the reconstrcutions I made of plant cells. http://home.nvg.ntnu.no/~gary/MSc/IML1.html Gary. >-----Original Message----- >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] >Sent: 29. oktober 1999 06:37 >To: nih-image@io.ece.drexel.edu >Subject: 3d reconstruction > >A colleague wishes to make slices through a frog's leg, examine them with a >microscope, then capture the images (using a Scion board and NIH-Image) and >reconstruct the images into a 3D model. Is there any software or Image macro >out >there that will help? >Thanks > >-- >jacques soddell >biological sciences, school of management technology and environment >la trobe university bendigo, po box 199, bendigo, victoria, australia. >j.soddell@bendigo.latrobe.edu.au >http://redgum.bendigo.latrobe.edu.au/~soddell >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 >"everything is music, everything is noise" (john cage) >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html > > -------------------------------- End of nih-image-d Digest V99 Issue #250 **************************************** From nih-image-request@io.ece.drexel.edu Tue Nov 2 05:36 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA24310; Tue, 2 Nov 1999 05:36:11 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 05:36:11 -0500 (EST) Message-ID: From: Gary Chinga To: "'nih-image@io.ece.drexel.edu'" Subject: RE: 3d reconstruction Date: Tue, 2 Nov 1999 11:18:35 +0100 X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.996.62 MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Resent-Message-ID: <"_mB5o.0.Nu4.6dh7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1898 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1073 Status: O You can use NIH Image or better Object Image to make reconstructions, take a look at this page and see the reconstrcutions I made of plant cells. http://home.nvg.ntnu.no/~gary/MSc/IML1.html Gary. >-----Original Message----- >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] >Sent: 29. oktober 1999 06:37 >To: nih-image@io.ece.drexel.edu >Subject: 3d reconstruction > >A colleague wishes to make slices through a frog's leg, examine them with a >microscope, then capture the images (using a Scion board and NIH-Image) and >reconstruct the images into a 3D model. Is there any software or Image macro >out >there that will help? >Thanks > >-- >jacques soddell >biological sciences, school of management technology and environment >la trobe university bendigo, po box 199, bendigo, victoria, australia. >j.soddell@bendigo.latrobe.edu.au >http://redgum.bendigo.latrobe.edu.au/~soddell >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 >"everything is music, everything is noise" (john cage) >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html > > From nih-image-request@io.ece.drexel.edu Tue Nov 2 07:08 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA10455; Tue, 2 Nov 1999 07:08:47 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 07:08:47 -0500 (EST) From: "Leonardo Borghi" To: , Subject: Unsubscribe Date: Mon, 1 Nov 1999 08:36:21 -0200 Message-ID: <01bf2454$f04d16e0$3995f4c8@osiris.ccard.com.br> MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 Resent-Message-ID: <"8UeAz3.0.Rf1.0-i7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1899 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; boundary="----------------------------"; charset="iso-8859-1" Content-Length: 2 Status: O From nih-image-request@io.ece.drexel.edu Tue Nov 2 07:47 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA17353; Tue, 2 Nov 1999 07:47:25 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 07:47:25 -0500 (EST) Message-Id: X-Mailer: Novell GroupWise 5.5.2 Date: Tue, 02 Nov 1999 07:31:33 -0500 From: "Tim Crowe" To: Subject: Re: nih-image-d Digest V99 #250 Mime-Version: 1.0 Content-Disposition: inline Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by io.ece.drexel.edu id HAA14565 Resent-Message-ID: <"MSyB51.0.oZ3.mdj7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1900 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=US-ASCII Content-Length: 114 Status: O ** Confidential ** ** High Priority ** Please remove my name from the mailing list. Thank you. crowet@ccf.org From nih-image-request@io.ece.drexel.edu Tue Nov 2 08:48 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA28430; Tue, 2 Nov 1999 08:48:53 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 08:48:53 -0500 (EST) Message-ID: <19991102133031.24832.rocketmail@web601.yahoomail.com> Date: Tue, 2 Nov 1999 05:30:31 -0800 (PST) From: steve spencer Subject: Re: Scion Image experience? To: nih-image@io.ece.drexel.edu MIME-Version: 1.0 Resent-Message-ID: <"EP7bg3.0.956.yTk7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1901 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 1155 Status: O scion image is still full of bugs too numerous to report here. I would suggest a good emulator so you can un the mac os on your pc. steve if not then try imagej. it is java based and platform independent --- Rachel Spicer wrote: > Hi all- > > I've just left a MAC-based lab that used NIH Image > and hope to set up a > similar imaging facility (currently PC-based) at my > new institution. I'm > wondering if anyone out there has experience with > the Scion Image program > available for the PC. I remember the beta versions > were full of bugs, and > I am curious how the current version compares to NIH > Image v. 1.61. Any > thoughts? > > Thanks in advance, > > Rachel > > ****************************************** > Rachel Spicer > Biological Laboratories 395 > Organismic and Evolutionary Biology > Harvard University > 16 Divinity Avenue > Cambridge, MA 02138 > > (617) 496-3580 (phone) > (617) 496-5854 (fax) > spicer@oeb.harvard.edu > ****************************************** > > ===== __________________________________________________ Do You Yahoo!? Bid and sell for free at http://auctions.yahoo.com From nih-image-request@io.ece.drexel.edu Tue Nov 2 11:11 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA25039; Tue, 2 Nov 1999 11:11:50 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 11:11:50 -0500 (EST) Message-ID: <381F0835.C49F1728@ucsd.edu> Date: Tue, 02 Nov 1999 07:50:13 -0800 From: Harvey Karten Organization: UCSD X-Mailer: Mozilla 4.04 [en] (Win95; U) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: 3D reconstruction References: <199911021029.FAA22931@io.ece.drexel.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"9I0Q22.0.425.XXm7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1902 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 1502 Status: O Gary, Very neat. What were your exact steps required to generate the images shown on your webpage? How did you generate the internal solid objects? regards, Harvey > > > Subject: RE: 3d reconstruction > Date: Tue, 2 Nov 1999 11:18:35 +0100 > From: Gary Chinga > To: "'nih-image@io.ece.drexel.edu'" > > You can use NIH Image or better Object Image to make reconstructions, > take a look at this page and see the reconstrcutions I made of plant > cells. > > http://home.nvg.ntnu.no/~gary/MSc/IML1.html > > Gary. > > >-----Original Message----- > >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] > >Sent: 29. oktober 1999 06:37 > >To: nih-image@io.ece.drexel.edu > >Subject: 3d reconstruction > > > >A colleague wishes to make slices through a frog's leg, examine them with a > >microscope, then capture the images (using a Scion board and NIH-Image) and > >reconstruct the images into a 3D model. Is there any software or Image macro > >out > >there that will help? > >Thanks > > > >-- > >jacques soddell > >biological sciences, school of management technology and environment > >la trobe university bendigo, po box 199, bendigo, victoria, australia. > >j.soddell@bendigo.latrobe.edu.au > >http://redgum.bendigo.latrobe.edu.au/~soddell > >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 > >"everything is music, everything is noise" (john cage) > >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html > > > > From nih-image-request@io.ece.drexel.edu Tue Nov 2 11:25 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA27629; Tue, 2 Nov 1999 11:25:07 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 11:25:07 -0500 (EST) X-Authentication-Warning: mail.bio.uva.nl: Host gold.bio.uva.nl [145.18.160.48] claimed to be [145.18.160.48] Message-Id: In-Reply-To: <22AA496BEC@rna.bio.mq.edu.au> Mime-Version: 1.0 Date: Tue, 2 Nov 1999 17:05:55 +0100 To: nih-image@io.ece.drexel.edu From: Norbert Vischer Subject: Re: Distance between lines Resent-Message-ID: <"RT0Fh2.0.8h5.1jm7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1903 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 759 Status: O Stephen, I have meanwhile studied Greg Joss' macro more thoroughly and realized what elegant approach it is. His solution directly addresses the min/mean/max results you need. When I tested the macro, I sometimes obtained skeletons with bifurkations, which disturbed the measurements - obviously round ends of the tube are required. If you prefer to work with non-destructive objects, his ideas could be combined with roi objects etc. N. Vischer Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html From nih-image-request@io.ece.drexel.edu Tue Nov 2 11:43 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id LAA01217; Tue, 2 Nov 1999 11:43:06 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 11:43:06 -0500 (EST) Message-ID: From: Gary Chinga To: "'nih-image@io.ece.drexel.edu'" Subject: RE: 3D reconstruction Date: Tue, 2 Nov 1999 17:25:21 +0100 X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.996.62 MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Resent-Message-ID: <"l23FR2.0.Cc6.z-m7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1904 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 2483 Status: O The whole process of sectioning, acquisition and processing of the images can become very tedious and time consuming. It is really important to use macros to automate some steps. So if you cant write macros, use some hours to learn the basics of the macro language NIH Image uses, it will save you for a lot of work later. I dont think it is possible to explain my whole procedure in just a few lines, but you can download my entire Master thesis from the same web page. http://home.nvg.ntnu.no/~gary/MSc/thesis.html Take a look at section 2 of chapter 3 in my thesis. You can also find some valuable information in chapter 2 regarding other software packages available at that time. Keep in mind that my thesis were written in 1995!!! Regards, Gary. >-----Original Message----- >From: Harvey Karten [SMTP:hjkarten@ucsd.edu] >Sent: 2. november 1999 16:50 >To: nih-image@io.ece.drexel.edu >Subject: 3D reconstruction > >Gary, Very neat. What were your exact steps required to generate the >images shown >on your webpage? How did you generate the internal solid objects? > >regards, Harvey > >> >> >> Subject: RE: 3d reconstruction >> Date: Tue, 2 Nov 1999 11:18:35 +0100 >> From: Gary Chinga >> To: "'nih-image@io.ece.drexel.edu'" >> >> You can use NIH Image or better Object Image to make reconstructions, >> take a look at this page and see the reconstrcutions I made of plant >> cells. >> >> http://home.nvg.ntnu.no/~gary/MSc/IML1.html >> >> Gary. >> >> >-----Original Message----- >> >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] >> >Sent: 29. oktober 1999 06:37 >> >To: nih-image@io.ece.drexel.edu >> >Subject: 3d reconstruction >> > >> >A colleague wishes to make slices through a frog's leg, examine them with >>a >> >microscope, then capture the images (using a Scion board and NIH-Image) >>and >> >reconstruct the images into a 3D model. Is there any software or Image >>macro >> >out >> >there that will help? >> >Thanks >> > >> >-- >> >jacques soddell >> >biological sciences, school of management technology and environment >> >la trobe university bendigo, po box 199, bendigo, victoria, australia. >> >j.soddell@bendigo.latrobe.edu.au >> >http://redgum.bendigo.latrobe.edu.au/~soddell >> >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 >> >"everything is music, everything is noise" (john cage) >> >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html >> > >> > > > > From nih-image-request@io.ece.drexel.edu Tue Nov 2 18:58 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA20009; Tue, 2 Nov 1999 18:58:39 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 18:58:39 -0500 (EST) From: J.Valenta@sh.cvut.cz Organization: SPS Strahov To: nih-image@io.ece.drexel.edu Date: Wed, 3 Nov 1999 00:38:04 MEST Subject: Unsubscribe Priority: normal X-mailer: Pegasus Mail v3.40 Message-ID: Resent-Message-ID: <"oZEvD1.0.4y3.hOt7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1905 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 751 Status: O Check out my www server: equs.sh.cvut.cz Check out my home page : www.cbmi.cvut.cz/aktivity/ultrazvuk/soft/segment/index1.html *************************************** * mail back to : j.valenta@sh.cvut.cz * * student-faculty of electrical eng. * * > cybernetics < * * CZECH TECHNICAL UNIVERZITY * *-------------------------------------* * eli , eli lama sabachtani * * morituri te salutant * * I have a dream ... * *************************************** From nih-image-request@io.ece.drexel.edu Tue Nov 2 19:09 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA22083; Tue, 2 Nov 1999 19:09:19 -0500 (EST) Resent-Date: Tue, 2 Nov 1999 19:09:19 -0500 (EST) From: "Kamla Ahluwalia [Anatomy]" Organization: SUNY Stony Brook To: nih-image@io.ece.drexel.edu Date: Tue, 2 Nov 1999 18:48:58 -0500 MIME-Version: 1.0 Content-transfer-encoding: 7BIT Subject: edge detection help needed Priority: normal X-mailer: Pegasus Mail for Windows (v2.54) Message-ID: <7945186259@informatics-2.informatics.sunysb.edu> Resent-Message-ID: <"wldGx2.0.9R4.SYt7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1906 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=US-ASCII Content-Length: 562 Status: O Hello I've recently started using Scion Image to analyse CT scans. I'm trying to identify the separation between the subchondral cortical and trabecular bone. Scion apparently uses a Sobel edge-detection technique, but it doesn't provide a satisfactory explanation of it's processes. The image looks good, but I want to know exactly what it's doing to determine the edge. Can anyone suggest a source for this information? Or perhaps provide me with a relatively simple explanation (n.b., I am neither a programmer nor a mathematician). Thank you, Kamla From nih-image-d-request@io.ece.drexel.edu Wed Nov 3 03:08 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id DAA13473; Wed, 3 Nov 1999 03:08:18 -0500 (EST) Date: Wed, 3 Nov 1999 03:08:18 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911030808.DAA13473@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #251 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/251 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 11287 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 251 Today's Topics: Unsubscribe [ "Leonardo Borghi" ] Re: Scion Image experience? [ steve spencer ] Re: Distance between lines [ Norbert Vischer ] Unsubscribe [ J.Valenta@sh.cvut.cz ] edge detection help needed [ "Kamla Ahluwalia [Anatomy]" To: , Subject: Unsubscribe Message-ID: <01bf2454$f04d16e0$3995f4c8@osiris.ccard.com.br> Content-Transfer-Encoding: 7bit Content-Type: text/plain; boundary="----------------------------"; charset="iso-8859-1" ------------------------------ Date: Tue, 02 Nov 1999 07:31:33 -0500 From: "Tim Crowe" To: Subject: Re: nih-image-d Digest V99 #250 Message-Id: Content-Type: text/plain; charset=US-ASCII Content-Disposition: inline Content-Transfer-Encoding: 8bit ** Confidential ** ** High Priority ** Please remove my name from the mailing list. Thank you. crowet@ccf.org ------------------------------ Date: Tue, 2 Nov 1999 05:30:31 -0800 (PST) From: steve spencer To: nih-image@io.ece.drexel.edu Subject: Re: Scion Image experience? Message-ID: <19991102133031.24832.rocketmail@web601.yahoomail.com> Content-Type: text/plain; charset=us-ascii scion image is still full of bugs too numerous to report here. I would suggest a good emulator so you can un the mac os on your pc. steve if not then try imagej. it is java based and platform independent --- Rachel Spicer wrote: > Hi all- > > I've just left a MAC-based lab that used NIH Image > and hope to set up a > similar imaging facility (currently PC-based) at my > new institution. I'm > wondering if anyone out there has experience with > the Scion Image program > available for the PC. I remember the beta versions > were full of bugs, and > I am curious how the current version compares to NIH > Image v. 1.61. Any > thoughts? > > Thanks in advance, > > Rachel > > ****************************************** > Rachel Spicer > Biological Laboratories 395 > Organismic and Evolutionary Biology > Harvard University > 16 Divinity Avenue > Cambridge, MA 02138 > > (617) 496-3580 (phone) > (617) 496-5854 (fax) > spicer@oeb.harvard.edu > ****************************************** > > ===== __________________________________________________ Do You Yahoo!? Bid and sell for free at http://auctions.yahoo.com ------------------------------ Date: Tue, 02 Nov 1999 07:50:13 -0800 From: Harvey Karten To: nih-image@io.ece.drexel.edu Subject: 3D reconstruction Message-ID: <381F0835.C49F1728@ucsd.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Gary, Very neat. What were your exact steps required to generate the images shown on your webpage? How did you generate the internal solid objects? regards, Harvey > > > Subject: RE: 3d reconstruction > Date: Tue, 2 Nov 1999 11:18:35 +0100 > From: Gary Chinga > To: "'nih-image@io.ece.drexel.edu'" > > You can use NIH Image or better Object Image to make reconstructions, > take a look at this page and see the reconstrcutions I made of plant > cells. > > http://home.nvg.ntnu.no/~gary/MSc/IML1.html > > Gary. > > >-----Original Message----- > >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] > >Sent: 29. oktober 1999 06:37 > >To: nih-image@io.ece.drexel.edu > >Subject: 3d reconstruction > > > >A colleague wishes to make slices through a frog's leg, examine them with a > >microscope, then capture the images (using a Scion board and NIH-Image) and > >reconstruct the images into a 3D model. Is there any software or Image macro > >out > >there that will help? > >Thanks > > > >-- > >jacques soddell > >biological sciences, school of management technology and environment > >la trobe university bendigo, po box 199, bendigo, victoria, australia. > >j.soddell@bendigo.latrobe.edu.au > >http://redgum.bendigo.latrobe.edu.au/~soddell > >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 > >"everything is music, everything is noise" (john cage) > >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html > > > > ------------------------------ Date: Tue, 2 Nov 1999 17:05:55 +0100 From: Norbert Vischer To: nih-image@io.ece.drexel.edu Subject: Re: Distance between lines Message-Id: Content-Type: text/plain; charset="us-ascii" Stephen, I have meanwhile studied Greg Joss' macro more thoroughly and realized what elegant approach it is. His solution directly addresses the min/mean/max results you need. When I tested the macro, I sometimes obtained skeletons with bifurkations, which disturbed the measurements - obviously round ends of the tube are required. If you prefer to work with non-destructive objects, his ideas could be combined with roi objects etc. N. Vischer Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html ------------------------------ Date: Tue, 2 Nov 1999 17:25:21 +0100 From: Gary Chinga To: "'nih-image@io.ece.drexel.edu'" Subject: RE: 3D reconstruction Message-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit The whole process of sectioning, acquisition and processing of the images can become very tedious and time consuming. It is really important to use macros to automate some steps. So if you cant write macros, use some hours to learn the basics of the macro language NIH Image uses, it will save you for a lot of work later. I dont think it is possible to explain my whole procedure in just a few lines, but you can download my entire Master thesis from the same web page. http://home.nvg.ntnu.no/~gary/MSc/thesis.html Take a look at section 2 of chapter 3 in my thesis. You can also find some valuable information in chapter 2 regarding other software packages available at that time. Keep in mind that my thesis were written in 1995!!! Regards, Gary. >-----Original Message----- >From: Harvey Karten [SMTP:hjkarten@ucsd.edu] >Sent: 2. november 1999 16:50 >To: nih-image@io.ece.drexel.edu >Subject: 3D reconstruction > >Gary, Very neat. What were your exact steps required to generate the >images shown >on your webpage? How did you generate the internal solid objects? > >regards, Harvey > >> >> >> Subject: RE: 3d reconstruction >> Date: Tue, 2 Nov 1999 11:18:35 +0100 >> From: Gary Chinga >> To: "'nih-image@io.ece.drexel.edu'" >> >> You can use NIH Image or better Object Image to make reconstructions, >> take a look at this page and see the reconstrcutions I made of plant >> cells. >> >> http://home.nvg.ntnu.no/~gary/MSc/IML1.html >> >> Gary. >> >> >-----Original Message----- >> >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] >> >Sent: 29. oktober 1999 06:37 >> >To: nih-image@io.ece.drexel.edu >> >Subject: 3d reconstruction >> > >> >A colleague wishes to make slices through a frog's leg, examine them with >>a >> >microscope, then capture the images (using a Scion board and NIH-Image) >>and >> >reconstruct the images into a 3D model. Is there any software or Image >>macro >> >out >> >there that will help? >> >Thanks >> > >> >-- >> >jacques soddell >> >biological sciences, school of management technology and environment >> >la trobe university bendigo, po box 199, bendigo, victoria, australia. >> >j.soddell@bendigo.latrobe.edu.au >> >http://redgum.bendigo.latrobe.edu.au/~soddell >> >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 >> >"everything is music, everything is noise" (john cage) >> >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html >> > >> > > > > ------------------------------ Date: Wed, 3 Nov 1999 00:38:04 MEST From: J.Valenta@sh.cvut.cz To: nih-image@io.ece.drexel.edu Subject: Unsubscribe Message-ID: Check out my www server: equs.sh.cvut.cz Check out my home page : www.cbmi.cvut.cz/aktivity/ultrazvuk/soft/segment/index1.html *************************************** * mail back to : j.valenta@sh.cvut.cz * * student-faculty of electrical eng. * * > cybernetics < * * CZECH TECHNICAL UNIVERZITY * *-------------------------------------* * eli , eli lama sabachtani * * morituri te salutant * * I have a dream ... * *************************************** ------------------------------ Date: Tue, 2 Nov 1999 18:48:58 -0500 From: "Kamla Ahluwalia [Anatomy]" To: nih-image@io.ece.drexel.edu Subject: edge detection help needed Message-ID: <7945186259@informatics-2.informatics.sunysb.edu> Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Hello I've recently started using Scion Image to analyse CT scans. I'm trying to identify the separation between the subchondral cortical and trabecular bone. Scion apparently uses a Sobel edge-detection technique, but it doesn't provide a satisfactory explanation of it's processes. The image looks good, but I want to know exactly what it's doing to determine the edge. Can anyone suggest a source for this information? Or perhaps provide me with a relatively simple explanation (n.b., I am neither a programmer nor a mathematician). Thank you, Kamla ------------------------------ Date: Wed, 03 Nov 1999 08:53:51 +0100 From: Jochen Hoffmann To: nih-image@io.ece.drexel.edu Subject: Unsubscribe Message-Id: <3.0.5.32.19991103085351.007a2100@popserver.uni-konstanz.de> Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit ---------------------------------------------------------------------------- --- Jochen Hoffmann c/o Prof. K. Mendgen Uni Konstanz Fakultät für Biologie LS für Phytopathologie Universitätsstrasse 10 78457 Konstanz Germany -------------------------------- End of nih-image-d Digest V99 Issue #251 **************************************** From nih-image-request@io.ece.drexel.edu Wed Nov 3 03:10 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id DAA14062; Wed, 3 Nov 1999 03:10:59 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 03:10:59 -0500 (EST) Message-Id: <3.0.5.32.19991103085351.007a2100@popserver.uni-konstanz.de> X-Sender: pop01325@popserver.uni-konstanz.de X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Wed, 03 Nov 1999 08:53:51 +0100 To: nih-image@io.ece.drexel.edu From: Jochen Hoffmann Subject: Unsubscribe Mime-Version: 1.0 Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by io.ece.drexel.edu id CAA10897 Resent-Message-ID: <"aYeTS1.0.Ug2.wf-7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1907 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="iso-8859-1" Content-Length: 224 Status: O ---------------------------------------------------------------------------- --- Jochen Hoffmann c/o Prof. K. Mendgen Uni Konstanz Fakultät für Biologie LS für Phytopathologie Universitätsstrasse 10 78457 Konstanz Germany From nih-image-request@io.ece.drexel.edu Wed Nov 3 03:15 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id DAA14932; Wed, 3 Nov 1999 03:15:05 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 03:15:05 -0500 (EST) From: gsirat@optimet.co.il Message-ID: <50AF7BBCE9ADD1118F3A006008AFAFA4201F49@BALOO> To: nih-image@io.ece.drexel.edu Subject: unsuscribe Date: Wed, 3 Nov 1999 09:58:24 +0200 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2448.0) Resent-Message-ID: <"Px_p7.0.3t2.kj-7u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1908 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="WINDOWS-1255" Content-Length: 2 Status: O From nih-image-request@io.ece.drexel.edu Wed Nov 3 05:19 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA05917; Wed, 3 Nov 1999 05:19:54 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 05:19:54 -0500 (EST) Date: Wed, 3 Nov 1999 05:05:00 -0500 (EST) From: nih-image-owner@io.ece.drexel.edu Message-Id: <199911031005.FAA03205@io.ece.drexel.edu> To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Resent-Message-ID: <"FsCdJ.0.Jo.EZ08u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1909 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 3743 Status: O NIH-image Mailing List Help --------------------------- nih-image-d-request@biomed.drexel.edu - Aministration for Digest nih-image-request@biomed.drexel.edu - Administation for List ********************************************************* * DO NOT SEND ADMINISTRATIVE COMMANDS TO THE LIST * ********************************************************* nih-image@biomed.drexel.edu - Sends mail to everyone on the list Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- From nih-image-request@io.ece.drexel.edu Wed Nov 3 06:15 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA15471; Wed, 3 Nov 1999 06:15:20 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 06:15:20 -0500 (EST) Message-ID: <3820163E.E5673900@lemming0.lem.uni-karlsruhe.de> Date: Wed, 03 Nov 1999 12:02:22 +0100 From: reznik X-Mailer: Mozilla 4.6 [de] (Win95; I) X-Accept-Language: de MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: unsubscribe References: <199911030755.CAA10959@io.ece.drexel.edu> Content-Transfer-Encoding: 8bit Resent-Message-ID: <"KLaTi3.0.573.1O18u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1910 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=iso-8859-1 Content-Length: 10751 Status: O nih-image-d-request@io.ece.drexel.edu schrieb: > Betreff: > > nih-image-d Digest Volume 99 : Issue 251 > > Today's Topics: > Unsubscribe [ "Leonardo Borghi" Re: nih-image-d Digest V99 #250 [ "Tim Crowe" ] > Re: Scion Image experience? [ steve spencer 3D reconstruction [ Harvey Karten ] > Re: Distance between lines [ Norbert Vischer RE: 3D reconstruction [ Gary Chinga ] > Unsubscribe [ J.Valenta@sh.cvut.cz ] > edge detection help needed [ "Kamla Ahluwalia [Anatomy]" Unsubscribe [ Jochen Hoffmann > ------------------------------------------------------------------------ > > Betreff: Unsubscribe > Datum: Mon, 1 Nov 1999 08:36:21 -0200 > Von: "Leonardo Borghi" > An: , > > Betreff: Re: nih-image-d Digest V99 #250 > Datum: Tue, 02 Nov 1999 07:31:33 -0500 > Von: "Tim Crowe" > An: > > ** Confidential ** > ** High Priority ** > > Please remove my name from the mailing list. > > Thank you. > > crowet@ccf.org > > ------------------------------------------------------------------------ > > Betreff: Re: Scion Image experience? > Datum: Tue, 2 Nov 1999 05:30:31 -0800 (PST) > Von: steve spencer > An: nih-image@io.ece.drexel.edu > > scion image is still full of bugs too numerous to > report here. I would suggest a good emulator so you > can un the mac os on your pc. > steve > if not then try imagej. it is java based and platform > independent > > --- Rachel Spicer wrote: > > Hi all- > > > > I've just left a MAC-based lab that used NIH Image > > and hope to set up a > > similar imaging facility (currently PC-based) at my > > new institution. I'm > > wondering if anyone out there has experience with > > the Scion Image program > > available for the PC. I remember the beta versions > > were full of bugs, and > > I am curious how the current version compares to NIH > > Image v. 1.61. Any > > thoughts? > > > > Thanks in advance, > > > > Rachel > > > > ****************************************** > > Rachel Spicer > > Biological Laboratories 395 > > Organismic and Evolutionary Biology > > Harvard University > > 16 Divinity Avenue > > Cambridge, MA 02138 > > > > (617) 496-3580 (phone) > > (617) 496-5854 (fax) > > spicer@oeb.harvard.edu > > ****************************************** > > > > > > ===== > > __________________________________________________ > Do You Yahoo!? > Bid and sell for free at http://auctions.yahoo.com > > ------------------------------------------------------------------------ > > Betreff: 3D reconstruction > Datum: Tue, 02 Nov 1999 07:50:13 -0800 > Von: Harvey Karten > An: nih-image@io.ece.drexel.edu > > Gary, Very neat. What were your exact steps required to generate the images shown > on your webpage? How did you generate the internal solid objects? > > regards, Harvey > > > > > > > Subject: RE: 3d reconstruction > > Date: Tue, 2 Nov 1999 11:18:35 +0100 > > From: Gary Chinga > > To: "'nih-image@io.ece.drexel.edu'" > > > > You can use NIH Image or better Object Image to make reconstructions, > > take a look at this page and see the reconstrcutions I made of plant > > cells. > > > > http://home.nvg.ntnu.no/~gary/MSc/IML1.html > > > > Gary. > > > > >-----Original Message----- > > >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] > > >Sent: 29. oktober 1999 06:37 > > >To: nih-image@io.ece.drexel.edu > > >Subject: 3d reconstruction > > > > > >A colleague wishes to make slices through a frog's leg, examine them with a > > >microscope, then capture the images (using a Scion board and NIH-Image) and > > >reconstruct the images into a 3D model. Is there any software or Image macro > > >out > > >there that will help? > > >Thanks > > > > > >-- > > >jacques soddell > > >biological sciences, school of management technology and environment > > >la trobe university bendigo, po box 199, bendigo, victoria, australia. > > >j.soddell@bendigo.latrobe.edu.au > > >http://redgum.bendigo.latrobe.edu.au/~soddell > > >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 > > >"everything is music, everything is noise" (john cage) > > >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html > > > > > > > > ------------------------------------------------------------------------ > > Betreff: Re: Distance between lines > Datum: Tue, 2 Nov 1999 17:05:55 +0100 > Von: Norbert Vischer > An: nih-image@io.ece.drexel.edu > > Stephen, > > I have meanwhile studied Greg Joss' macro more thoroughly and realized > what elegant approach it is. His solution directly addresses the > min/mean/max results you need. > When I tested the macro, I sometimes obtained skeletons with bifurkations, > which disturbed the measurements - obviously round ends of the tube are > required. If you prefer to work with non-destructive objects, his ideas > could be combined with roi objects etc. > > N. Vischer > > Norbert Vischer University of Amsterdam > scientific engineer Molecular Cell Biology > Kruislaan 316 > tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam > e-mail: vischer@bio.uva.nl The Netherlands > http://simon.bio.uva.nl/object-image.html > > ------------------------------------------------------------------------ > > Betreff: RE: 3D reconstruction > Datum: Tue, 2 Nov 1999 17:25:21 +0100 > Von: Gary Chinga > An: "'nih-image@io.ece.drexel.edu'" > > The whole process of sectioning, acquisition and processing of the > images can become very tedious and time consuming. It is really > important to use macros to automate some steps. So if you cant write > macros, use some hours to learn the basics of the macro language NIH > Image uses, it will save you for a lot of work later. > > I dont think it is possible to explain my whole procedure in just a few > lines, but you can download my entire Master thesis from the same web > page. > > http://home.nvg.ntnu.no/~gary/MSc/thesis.html > > Take a look at section 2 of chapter 3 in my thesis. You can also find > some valuable information in chapter 2 regarding other software packages > available at that time. Keep in mind that my thesis were written in > 1995!!! > > Regards, > > Gary. > > >-----Original Message----- > >From: Harvey Karten [SMTP:hjkarten@ucsd.edu] > >Sent: 2. november 1999 16:50 > >To: nih-image@io.ece.drexel.edu > >Subject: 3D reconstruction > > > >Gary, Very neat. What were your exact steps required to generate the > >images shown > >on your webpage? How did you generate the internal solid objects? > > > >regards, Harvey > > > >> > >> > >> Subject: RE: 3d reconstruction > >> Date: Tue, 2 Nov 1999 11:18:35 +0100 > >> From: Gary Chinga > >> To: "'nih-image@io.ece.drexel.edu'" > >> > >> You can use NIH Image or better Object Image to make reconstructions, > >> take a look at this page and see the reconstrcutions I made of plant > >> cells. > >> > >> http://home.nvg.ntnu.no/~gary/MSc/IML1.html > >> > >> Gary. > >> > >> >-----Original Message----- > >> >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] > >> >Sent: 29. oktober 1999 06:37 > >> >To: nih-image@io.ece.drexel.edu > >> >Subject: 3d reconstruction > >> > > >> >A colleague wishes to make slices through a frog's leg, examine them with > >>a > >> >microscope, then capture the images (using a Scion board and NIH-Image) > >>and > >> >reconstruct the images into a 3D model. Is there any software or Image > >>macro > >> >out > >> >there that will help? > >> >Thanks > >> > > >> >-- > >> >jacques soddell > >> >biological sciences, school of management technology and environment > >> >la trobe university bendigo, po box 199, bendigo, victoria, australia. > >> >j.soddell@bendigo.latrobe.edu.au > >> >http://redgum.bendigo.latrobe.edu.au/~soddell > >> >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 > >> >"everything is music, everything is noise" (john cage) > >> >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html > >> > > >> > > > > > > > > > ------------------------------------------------------------------------ > > Betreff: Unsubscribe > Datum: Wed, 3 Nov 1999 00:38:04 MEST > Von: J.Valenta@sh.cvut.cz > An: nih-image@io.ece.drexel.edu > > Check out my www server: > equs.sh.cvut.cz > > Check out my home page : > www.cbmi.cvut.cz/aktivity/ultrazvuk/soft/segment/index1.html > > *************************************** > * mail back to : j.valenta@sh.cvut.cz * > * student-faculty of electrical eng. * > * > cybernetics < * > * CZECH TECHNICAL UNIVERZITY * > *-------------------------------------* > * eli , eli lama sabachtani * > * morituri te salutant * > * I have a dream ... * > *************************************** > > ------------------------------------------------------------------------ > > Betreff: edge detection help needed > Datum: Tue, 2 Nov 1999 18:48:58 -0500 > Von: "Kamla Ahluwalia [Anatomy]" > An: nih-image@io.ece.drexel.edu > > Hello > I've recently started using Scion Image to analyse CT scans. I'm > trying to identify the separation between the subchondral cortical > and trabecular bone. Scion apparently uses a Sobel edge-detection > technique, but it doesn't provide a satisfactory explanation of it's > processes. The image looks good, but I want to know exactly what it's > doing to determine the edge. Can anyone suggest a source for this > information? Or perhaps provide me with a relatively simple > explanation (n.b., I am neither a programmer nor a mathematician). > > Thank you, > Kamla > > ------------------------------------------------------------------------ > > Betreff: Unsubscribe > Datum: Wed, 03 Nov 1999 08:53:51 +0100 > Von: Jochen Hoffmann > An: nih-image@io.ece.drexel.edu > > ---------------------------------------------------------------------------- > --- > Jochen Hoffmann > c/o Prof. K. Mendgen > Uni Konstanz > Fakultät für Biologie > LS für Phytopathologie > Universitätsstrasse 10 > 78457 Konstanz > Germany From nih-image-request@io.ece.drexel.edu Wed Nov 3 06:16 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA15681; Wed, 3 Nov 1999 06:16:06 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 06:16:06 -0500 (EST) Message-ID: <382016A3.8790B122@lemming0.lem.uni-karlsruhe.de> Date: Wed, 03 Nov 1999 12:04:03 +0100 From: reznik X-Mailer: Mozilla 4.6 [de] (Win95; I) X-Accept-Language: de MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: Unsubscribe References: Content-Transfer-Encoding: 7bit Resent-Message-ID: <"rJpcE1.0.qB3.RP18u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1911 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=koi8-r Content-Length: 3 Status: O From nih-image-request@io.ece.drexel.edu Wed Nov 3 15:50 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id PAA25539; Wed, 3 Nov 1999 15:50:46 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 15:50:46 -0500 (EST) Message-Id: <199911032026.PAA20306@post-ofc05.srv.cis.pitt.edu> Date: Wed, 03 Nov 1999 16:25:12 -0400 From: Anu Gupta X-Mailer: Mozilla 4.5 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: Publishing/publicising macros References: <19991019132004.88972.qmail@hotmail.com> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"6VXhN2.0.AJ5.7j98u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1912 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Length: 213 Status: O i have a copper film with holes and I need to measure the perimeter of the holes however when I threshold and analyze, a get a count of 1, no reading of perimeter. Is there a way to do this? thanks -Anu Gupta From nih-image-request@io.ece.drexel.edu Wed Nov 3 17:02 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id RAA07801; Wed, 3 Nov 1999 17:02:29 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 17:02:29 -0500 (EST) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: angst16+@pitt.edu, nih-image@io.ece.drexel.edu Date: Thu, 4 Nov 1999 8:48:27 +1100 Subject: Re: Publishing/publicising macros Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <677E94307B@rna.bio.mq.edu.au> Resent-Message-ID: <"NpzE4.0.n21.hnA8u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1913 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 750 Status: O >Date: Wed, 03 Nov 1999 16:25:12 -0400 >From: Anu Gupta >To: nih-image@io.ece.drexel.edu >Subject: Re: Publishing/publicising macros > >i have a copper film with holes and I need to measure the perimeter of >the holes however when I threshold and analyze, a get a count of 1, no >reading of perimeter. Is there a way to do this? > >thanks >-Anu Gupta > I suspect that you need to 'invert(edit menu' your image so that threshold gives the holes as black particles on a white background (rather than one big black particle with holes in it). Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Wed Nov 3 17:47 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id RAA15510; Wed, 3 Nov 1999 17:47:33 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 17:47:33 -0500 (EST) From: "R. Aaron Falk" To: Subject: Contrast and Brightness Date: Wed, 3 Nov 1999 14:31:21 -0800 Message-ID: <002f01bf264b$27839240$0200a8c0@internet> MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.3612.1700 Importance: Normal Resent-Message-ID: <"ikXTw1.0.Mz2.WTB8u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1914 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="iso-8859-1" Content-Length: 2645 Status: O Hi all, Ever have one of those questions that you thought was obvious until you tried to answer it in detail. The question I had was "How does NIH Image change the palette in response to the brightness and contrast controls". Sounds obvious, but then some of the following issues came up. *** As a starting assumption for gray scale images *** Brightness implies that a constant is added to the baseline grayscale palette values. Contrast implies a change in slope, i.e. the baseline palette values are multiplied by the contrast slope. The modified gray palette values should be -- new = slope * original + constant -- where the original values for a gray scale run from 0-255, the slope is the contrast control (values from zero to infinity) and the constant is the brightness control (values from -255 to +255). This is the obvious part. Here are some of the sticky details. 1) Although I am reasonably certain that the NIH Image brightness and contrast controls operate as indicated above, I have been shown other programming texts that use different formulas for affecting contrast and brightness. Is there a standard definition of contrast and brightness in digital image processing? 2) As the contrast control is changed, the brightness level in NIH Image is seen to change slightly. The above assumptions would indicate that the two parameters should be independent. Why is this not the case? 3) In a grayscale image all the red, green, blue (RGB) values are set equal. Thus, the formula given above is simply applied to each of the separate color values. When a false color palette is used, the mapping between the above formula and the RGB values is rather unclear. NIH Image does what I would intuitively expect, but how does it actually handling this mapping? 4) The point of theses details has to do with opening a TIFF file saved from NIH Image in another application. Somehow, when NIH Image opens an image that it created, it is able to restore the brightness and contrast settings for that image - even with false color palettes. However, the standard TIFF file format only contains a palette - brightness and contrast are not part of standard. Thus, other applications with brightness and contrast controls do not recover the values set in NIH Image. How, exactly does NIH Image recover the brightness and contrast values, especially if a false color palette is being used? Are these values being stored somewhere in the TIFF file? All responses are greatly appreciated. Best regards, Aaron **************************************************************************** Dr. R. Aaron Falk President OptoMetrix, Inc. From nih-image-d-request@io.ece.drexel.edu Wed Nov 3 20:15 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id UAA13228; Wed, 3 Nov 1999 20:15:15 -0500 (EST) Date: Wed, 3 Nov 1999 20:15:15 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911040115.UAA13228@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #252 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/252 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 21738 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 252 Today's Topics: unsuscribe [ gsirat@optimet.co.il ] ADMIN: list commands [ nih-image-owner@io.ece.drexel.edu ] Re: unsubscribe [ reznik ] Re: Publishing/publicising macros [ GJOSS@rna.bio.mq.edu.au ] Contrast and Brightness [ "R. Aaron Falk" Content-Type: text/plain; charset="WINDOWS-1255" ------------------------------ Date: Wed, 3 Nov 1999 05:05:00 -0500 (EST) From: nih-image-owner@io.ece.drexel.edu To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Message-Id: <199911031005.FAA03205@io.ece.drexel.edu> NIH-image Mailing List Help --------------------------- nih-image-d-request@biomed.drexel.edu - Aministration for Digest nih-image-request@biomed.drexel.edu - Administation for List ********************************************************* * DO NOT SEND ADMINISTRATIVE COMMANDS TO THE LIST * ********************************************************* nih-image@biomed.drexel.edu - Sends mail to everyone on the list Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- ------------------------------ Date: Wed, 03 Nov 1999 12:02:22 +0100 From: reznik To: nih-image@io.ece.drexel.edu Subject: Re: unsubscribe Message-ID: <3820163E.E5673900@lemming0.lem.uni-karlsruhe.de> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit nih-image-d-request@io.ece.drexel.edu schrieb: > Betreff: > > nih-image-d Digest Volume 99 : Issue 251 > > Today's Topics: > Unsubscribe [ "Leonardo Borghi" Re: nih-image-d Digest V99 #250 [ "Tim Crowe" ] > Re: Scion Image experience? [ steve spencer 3D reconstruction [ Harvey Karten ] > Re: Distance between lines [ Norbert Vischer RE: 3D reconstruction [ Gary Chinga ] > Unsubscribe [ J.Valenta@sh.cvut.cz ] > edge detection help needed [ "Kamla Ahluwalia [Anatomy]" Unsubscribe [ Jochen Hoffmann > ------------------------------------------------------------------------ > > Betreff: Unsubscribe > Datum: Mon, 1 Nov 1999 08:36:21 -0200 > Von: "Leonardo Borghi" > An: , > > Betreff: Re: nih-image-d Digest V99 #250 > Datum: Tue, 02 Nov 1999 07:31:33 -0500 > Von: "Tim Crowe" > An: > > ** Confidential ** > ** High Priority ** > > Please remove my name from the mailing list. > > Thank you. > > crowet@ccf.org > > ------------------------------------------------------------------------ > > Betreff: Re: Scion Image experience? > Datum: Tue, 2 Nov 1999 05:30:31 -0800 (PST) > Von: steve spencer > An: nih-image@io.ece.drexel.edu > > scion image is still full of bugs too numerous to > report here. I would suggest a good emulator so you > can un the mac os on your pc. > steve > if not then try imagej. it is java based and platform > independent > > --- Rachel Spicer wrote: > > Hi all- > > > > I've just left a MAC-based lab that used NIH Image > > and hope to set up a > > similar imaging facility (currently PC-based) at my > > new institution. I'm > > wondering if anyone out there has experience with > > the Scion Image program > > available for the PC. I remember the beta versions > > were full of bugs, and > > I am curious how the current version compares to NIH > > Image v. 1.61. Any > > thoughts? > > > > Thanks in advance, > > > > Rachel > > > > ****************************************** > > Rachel Spicer > > Biological Laboratories 395 > > Organismic and Evolutionary Biology > > Harvard University > > 16 Divinity Avenue > > Cambridge, MA 02138 > > > > (617) 496-3580 (phone) > > (617) 496-5854 (fax) > > spicer@oeb.harvard.edu > > ****************************************** > > > > > > ===== > > __________________________________________________ > Do You Yahoo!? > Bid and sell for free at http://auctions.yahoo.com > > ------------------------------------------------------------------------ > > Betreff: 3D reconstruction > Datum: Tue, 02 Nov 1999 07:50:13 -0800 > Von: Harvey Karten > An: nih-image@io.ece.drexel.edu > > Gary, Very neat. What were your exact steps required to generate the images shown > on your webpage? How did you generate the internal solid objects? > > regards, Harvey > > > > > > > Subject: RE: 3d reconstruction > > Date: Tue, 2 Nov 1999 11:18:35 +0100 > > From: Gary Chinga > > To: "'nih-image@io.ece.drexel.edu'" > > > > You can use NIH Image or better Object Image to make reconstructions, > > take a look at this page and see the reconstrcutions I made of plant > > cells. > > > > http://home.nvg.ntnu.no/~gary/MSc/IML1.html > > > > Gary. > > > > >-----Original Message----- > > >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] > > >Sent: 29. oktober 1999 06:37 > > >To: nih-image@io.ece.drexel.edu > > >Subject: 3d reconstruction > > > > > >A colleague wishes to make slices through a frog's leg, examine them with a > > >microscope, then capture the images (using a Scion board and NIH-Image) and > > >reconstruct the images into a 3D model. Is there any software or Image macro > > >out > > >there that will help? > > >Thanks > > > > > >-- > > >jacques soddell > > >biological sciences, school of management technology and environment > > >la trobe university bendigo, po box 199, bendigo, victoria, australia. > > >j.soddell@bendigo.latrobe.edu.au > > >http://redgum.bendigo.latrobe.edu.au/~soddell > > >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 > > >"everything is music, everything is noise" (john cage) > > >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html > > > > > > > > ------------------------------------------------------------------------ > > Betreff: Re: Distance between lines > Datum: Tue, 2 Nov 1999 17:05:55 +0100 > Von: Norbert Vischer > An: nih-image@io.ece.drexel.edu > > Stephen, > > I have meanwhile studied Greg Joss' macro more thoroughly and realized > what elegant approach it is. His solution directly addresses the > min/mean/max results you need. > When I tested the macro, I sometimes obtained skeletons with bifurkations, > which disturbed the measurements - obviously round ends of the tube are > required. If you prefer to work with non-destructive objects, his ideas > could be combined with roi objects etc. > > N. Vischer > > Norbert Vischer University of Amsterdam > scientific engineer Molecular Cell Biology > Kruislaan 316 > tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam > e-mail: vischer@bio.uva.nl The Netherlands > http://simon.bio.uva.nl/object-image.html > > ------------------------------------------------------------------------ > > Betreff: RE: 3D reconstruction > Datum: Tue, 2 Nov 1999 17:25:21 +0100 > Von: Gary Chinga > An: "'nih-image@io.ece.drexel.edu'" > > The whole process of sectioning, acquisition and processing of the > images can become very tedious and time consuming. It is really > important to use macros to automate some steps. So if you cant write > macros, use some hours to learn the basics of the macro language NIH > Image uses, it will save you for a lot of work later. > > I dont think it is possible to explain my whole procedure in just a few > lines, but you can download my entire Master thesis from the same web > page. > > http://home.nvg.ntnu.no/~gary/MSc/thesis.html > > Take a look at section 2 of chapter 3 in my thesis. You can also find > some valuable information in chapter 2 regarding other software packages > available at that time. Keep in mind that my thesis were written in > 1995!!! > > Regards, > > Gary. > > >-----Original Message----- > >From: Harvey Karten [SMTP:hjkarten@ucsd.edu] > >Sent: 2. november 1999 16:50 > >To: nih-image@io.ece.drexel.edu > >Subject: 3D reconstruction > > > >Gary, Very neat. What were your exact steps required to generate the > >images shown > >on your webpage? How did you generate the internal solid objects? > > > >regards, Harvey > > > >> > >> > >> Subject: RE: 3d reconstruction > >> Date: Tue, 2 Nov 1999 11:18:35 +0100 > >> From: Gary Chinga > >> To: "'nih-image@io.ece.drexel.edu'" > >> > >> You can use NIH Image or better Object Image to make reconstructions, > >> take a look at this page and see the reconstrcutions I made of plant > >> cells. > >> > >> http://home.nvg.ntnu.no/~gary/MSc/IML1.html > >> > >> Gary. > >> > >> >-----Original Message----- > >> >From: Jacques Soddell [SMTP:J.Soddell@bendigo.latrobe.edu.au] > >> >Sent: 29. oktober 1999 06:37 > >> >To: nih-image@io.ece.drexel.edu > >> >Subject: 3d reconstruction > >> > > >> >A colleague wishes to make slices through a frog's leg, examine them with > >>a > >> >microscope, then capture the images (using a Scion board and NIH-Image) > >>and > >> >reconstruct the images into a 3D model. Is there any software or Image > >>macro > >> >out > >> >there that will help? > >> >Thanks > >> > > >> >-- > >> >jacques soddell > >> >biological sciences, school of management technology and environment > >> >la trobe university bendigo, po box 199, bendigo, victoria, australia. > >> >j.soddell@bendigo.latrobe.edu.au > >> >http://redgum.bendigo.latrobe.edu.au/~soddell > >> >possible musics, wednesday 10:30pm-midnight, 3ccc-fm 89.5 > >> >"everything is music, everything is noise" (john cage) > >> >http://redgum.bendigo.latrobe.edu.au/~soddell/pm.html > >> > > >> > > > > > > > > > ------------------------------------------------------------------------ > > Betreff: Unsubscribe > Datum: Wed, 3 Nov 1999 00:38:04 MEST > Von: J.Valenta@sh.cvut.cz > An: nih-image@io.ece.drexel.edu > > Check out my www server: > equs.sh.cvut.cz > > Check out my home page : > www.cbmi.cvut.cz/aktivity/ultrazvuk/soft/segment/index1.html > > *************************************** > * mail back to : j.valenta@sh.cvut.cz * > * student-faculty of electrical eng. * > * > cybernetics < * > * CZECH TECHNICAL UNIVERZITY * > *-------------------------------------* > * eli , eli lama sabachtani * > * morituri te salutant * > * I have a dream ... * > *************************************** > > ------------------------------------------------------------------------ > > Betreff: edge detection help needed > Datum: Tue, 2 Nov 1999 18:48:58 -0500 > Von: "Kamla Ahluwalia [Anatomy]" > An: nih-image@io.ece.drexel.edu > > Hello > I've recently started using Scion Image to analyse CT scans. I'm > trying to identify the separation between the subchondral cortical > and trabecular bone. Scion apparently uses a Sobel edge-detection > technique, but it doesn't provide a satisfactory explanation of it's > processes. The image looks good, but I want to know exactly what it's > doing to determine the edge. Can anyone suggest a source for this > information? Or perhaps provide me with a relatively simple > explanation (n.b., I am neither a programmer nor a mathematician). > > Thank you, > Kamla > > ------------------------------------------------------------------------ > > Betreff: Unsubscribe > Datum: Wed, 03 Nov 1999 08:53:51 +0100 > Von: Jochen Hoffmann > An: nih-image@io.ece.drexel.edu > > ---------------------------------------------------------------------------- > --- > Jochen Hoffmann > c/o Prof. K. Mendgen > Uni Konstanz > Fakultät für Biologie > LS für Phytopathologie > Universitätsstrasse 10 > 78457 Konstanz > Germany ------------------------------ Date: Wed, 03 Nov 1999 12:04:03 +0100 From: reznik To: nih-image@io.ece.drexel.edu Subject: Re: Unsubscribe Message-ID: <382016A3.8790B122@lemming0.lem.uni-karlsruhe.de> Content-Type: text/plain; charset=koi8-r Content-Transfer-Encoding: 7bit ------------------------------ Date: Wed, 03 Nov 1999 16:25:12 -0400 From: Anu Gupta To: nih-image@io.ece.drexel.edu Subject: Re: Publishing/publicising macros Message-Id: <199911032026.PAA20306@post-ofc05.srv.cis.pitt.edu> Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit i have a copper film with holes and I need to measure the perimeter of the holes however when I threshold and analyze, a get a count of 1, no reading of perimeter. Is there a way to do this? thanks -Anu Gupta ------------------------------ Date: Thu, 4 Nov 1999 8:48:27 +1100 From: GJOSS@rna.bio.mq.edu.au To: angst16+@pitt.edu, nih-image@io.ece.drexel.edu Subject: Re: Publishing/publicising macros Message-ID: <677E94307B@rna.bio.mq.edu.au> >Date: Wed, 03 Nov 1999 16:25:12 -0400 >From: Anu Gupta >To: nih-image@io.ece.drexel.edu >Subject: Re: Publishing/publicising macros > >i have a copper film with holes and I need to measure the perimeter of >the holes however when I threshold and analyze, a get a count of 1, no >reading of perimeter. Is there a way to do this? > >thanks >-Anu Gupta > I suspect that you need to 'invert(edit menu' your image so that threshold gives the holes as black particles on a white background (rather than one big black particle with holes in it). Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Wed, 3 Nov 1999 14:31:21 -0800 From: "R. Aaron Falk" To: Subject: Contrast and Brightness Message-ID: <002f01bf264b$27839240$0200a8c0@internet> Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Hi all, Ever have one of those questions that you thought was obvious until you tried to answer it in detail. The question I had was "How does NIH Image change the palette in response to the brightness and contrast controls". Sounds obvious, but then some of the following issues came up. *** As a starting assumption for gray scale images *** Brightness implies that a constant is added to the baseline grayscale palette values. Contrast implies a change in slope, i.e. the baseline palette values are multiplied by the contrast slope. The modified gray palette values should be -- new = slope * original + constant -- where the original values for a gray scale run from 0-255, the slope is the contrast control (values from zero to infinity) and the constant is the brightness control (values from -255 to +255). This is the obvious part. Here are some of the sticky details. 1) Although I am reasonably certain that the NIH Image brightness and contrast controls operate as indicated above, I have been shown other programming texts that use different formulas for affecting contrast and brightness. Is there a standard definition of contrast and brightness in digital image processing? 2) As the contrast control is changed, the brightness level in NIH Image is seen to change slightly. The above assumptions would indicate that the two parameters should be independent. Why is this not the case? 3) In a grayscale image all the red, green, blue (RGB) values are set equal. Thus, the formula given above is simply applied to each of the separate color values. When a false color palette is used, the mapping between the above formula and the RGB values is rather unclear. NIH Image does what I would intuitively expect, but how does it actually handling this mapping? 4) The point of theses details has to do with opening a TIFF file saved from NIH Image in another application. Somehow, when NIH Image opens an image that it created, it is able to restore the brightness and contrast settings for that image - even with false color palettes. However, the standard TIFF file format only contains a palette - brightness and contrast are not part of standard. Thus, other applications with brightness and contrast controls do not recover the values set in NIH Image. How, exactly does NIH Image recover the brightness and contrast values, especially if a false color palette is being used? Are these values being stored somewhere in the TIFF file? All responses are greatly appreciated. Best regards, Aaron **************************************************************************** Dr. R. Aaron Falk President OptoMetrix, Inc. ------------------------------ Date: Wed, 03 Nov 1999 16:59:38 -0800 From: Pat and Leslie Pringle To: NIHlist Subject: "survey mode" in Object Image Message-ID: <3820DA79.BC48030F@thurston.com> Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: 8bit There are times when 
"survey mode" (command key—space bar) does not display the entire image in a fit-to-screen mode in Object Image. I suspect that this may be a function of memory, because it typically happens when I am working on fairly sizeable files—any thoughts?
Thanks,
Pat Pringle

--
"Science is a way of trying not to fool yourself." (Richard Feynman)
  -------------------------------- End of nih-image-d Digest V99 Issue #252 **************************************** From nih-image-request@io.ece.drexel.edu Wed Nov 3 20:15 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id UAA13269; Wed, 3 Nov 1999 20:15:28 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 20:15:28 -0500 (EST) Message-ID: <3820DA79.BC48030F@thurston.com> Date: Wed, 03 Nov 1999 16:59:38 -0800 From: Pat and Leslie Pringle Reply-To: lespat@thurston.com X-Mailer: Mozilla 4.05 (Macintosh; I; PPC) MIME-Version: 1.0 To: NIHlist Subject: "survey mode" in Object Image Content-Transfer-Encoding: 8bit Resent-Message-ID: <"9MN421.0.mO2.1gD8u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1915 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/html; charset=iso-8859-1 Content-Length: 413 Status: O There are times when 
"survey mode" (command key—space bar) does not display the entire image in a fit-to-screen mode in Object Image. I suspect that this may be a function of memory, because it typically happens when I am working on fairly sizeable files—any thoughts?
Thanks,
Pat Pringle

--
"Science is a way of trying not to fool yourself." (Richard Feynman)
  From nih-image-request@io.ece.drexel.edu Wed Nov 3 23:35 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id XAA21970; Wed, 3 Nov 1999 23:35:53 -0500 (EST) Resent-Date: Wed, 3 Nov 1999 23:35:53 -0500 (EST) Date: Wed, 3 Nov 1999 23:16:10 -0500 (EST) From: Rajat Chakraborti To: nih-image@io.ece.drexel.edu Subject: Measurement options In-Reply-To: <677E94307B@rna.bio.mq.edu.au> Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"zy1VC2.0.NQ4.FYG8u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1916 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 998 Status: O I have been using NIH image software for analyzing suspended particle morphology. I would like to know about the exact measurements the software does. It will be of great help if anybody knows about it. I am asking in the "analyze" menu what are the exact measurements and what are the calculation from the measurement. Out of options of area, perimeter, S.D., major and minor side, angle etc., I think the major and minor side of aggregates are calculated from the measured area as a best fitted ellips (it is not measured actually). Is it correct? What is the source for looking the real measurements done by the software. ******************************************************************* Rajat Kanti Chakraborti State University of New York at Buffalo 207 Jarvis Hall Department of Civil, Structural and Environmental Engineering Buffalo, NY 14260 Ph. (716) 645-2114 ext. 2336 Fax: (716) 645-3667 rkc@acsu.buffalo.edu ******************************************************************** From nih-image-request@io.ece.drexel.edu Thu Nov 4 07:20 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA20215; Thu, 4 Nov 1999 07:20:51 -0500 (EST) Resent-Date: Thu, 4 Nov 1999 07:20:51 -0500 (EST) Date: Thu, 04 Nov 1999 11:59:28 +0000 From: "Dr. Patrick Felle" Subject: NIH Image and automated image analysis To: nih-image@io.ece.drexel.edu Reply-to: patrick.felle@ucd.ie Message-id: <3821751D.596C4587@ucd.ie> Organization: University College Dublin MIME-version: 1.0 X-Mailer: Mozilla 4.5 (Macintosh; I; PPC) X-Accept-Language: en-GB,fr-FR,Irish Resent-Message-ID: <"tn1p3.0.8z3.fLN8u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1917 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: multipart/alternative; boundary="------------CEF46AD17707635285885EB5" Content-Length: 1917 Status: O --------------CEF46AD17707635285885EB5 Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit I have been using a PC-based imaging system (Optimas running on NT, connected to automated stage light microscope). Will NIH Image run such a system on Macintosh G3 platform, if so what Image Grabber and Automatged stage is recommended? -- ----------------------------------- ----------------------------------- Dr. Patrick Felle (Director) Centre for Healthcare Informatics University College Dublin Earlsfort Terrace Dublin 2 IRELAND Tel / Fax: +353-1-706-7225 Mobile: 086 825 6045 --------------CEF46AD17707635285885EB5 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit I have been using a PC-based imaging system (Optimas running on NT, connected to automated stage light microscope).  Will NIH Image run such a system on Macintosh G3 platform, if so what Image Grabber and Automatged stage is recommended?
--



Dr. Patrick Felle (Director)
Centre for Healthcare Informatics
University College Dublin
Earlsfort Terrace
Dublin 2          IRELAND
Tel / Fax:  +353-1-706-7225     Mobile:  086 825 6045
--------------CEF46AD17707635285885EB5-- From nih-image-request@io.ece.drexel.edu Thu Nov 4 09:11 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id JAA12388; Thu, 4 Nov 1999 09:11:35 -0500 (EST) Resent-Date: Thu, 4 Nov 1999 09:11:35 -0500 (EST) X-Authentication-Warning: mail.bio.uva.nl: Host gold.bio.uva.nl [145.18.160.48] claimed to be [145.18.160.48] Message-Id: Mime-Version: 1.0 Date: Thu, 4 Nov 1999 14:55:44 +0100 To: nih-image@io.ece.drexel.edu From: Norbert Vischer Subject: Re: "survey mode" in Object Image Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by io.ece.drexel.edu id IAA08227 Resent-Message-ID: <"UJo6S3.0.t02.y-O8u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1918 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="iso-8859-1" Content-Length: 1176 Status: O > There are times when > "survey mode" (command key—space bar) does not display the entire image >in a fit-to-screen mode in Object Image. I suspect that this may be a >function of memory, because it typically happens when I am working on >fairly sizeable files—any thoughts? >Thanks, >Pat Pringle > >-- >"Science is a way of trying not to fool yourself." (Richard Feynman) > "Survey mode" (Command-SpaceBar) should temporarily have the same effect as clicking in the 'fit to screen' box on the right hand side of the title bar, except that a symbolized, movable window frame and a magnification menu become visible. Do you see any further difference? Do you use multiple monitors? Which version do you use? FitToScreen uses nearest neighbor scaling, i.e. small structures may disappear completely when they fall between the pixel grid due to extreme scaling-down. Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html From nih-image-d-request@io.ece.drexel.edu Fri Nov 5 02:18 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id CAA23514; Fri, 5 Nov 1999 02:18:17 -0500 (EST) Date: Fri, 5 Nov 1999 02:18:17 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911050718.CAA23514@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #253 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/253 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 6134 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 253 Today's Topics: Measurement options [ Rajat Chakraborti To: nih-image@io.ece.drexel.edu Subject: Measurement options Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII I have been using NIH image software for analyzing suspended particle morphology. I would like to know about the exact measurements the software does. It will be of great help if anybody knows about it. I am asking in the "analyze" menu what are the exact measurements and what are the calculation from the measurement. Out of options of area, perimeter, S.D., major and minor side, angle etc., I think the major and minor side of aggregates are calculated from the measured area as a best fitted ellips (it is not measured actually). Is it correct? What is the source for looking the real measurements done by the software. ******************************************************************* Rajat Kanti Chakraborti State University of New York at Buffalo 207 Jarvis Hall Department of Civil, Structural and Environmental Engineering Buffalo, NY 14260 Ph. (716) 645-2114 ext. 2336 Fax: (716) 645-3667 rkc@acsu.buffalo.edu ******************************************************************** ------------------------------ Date: Thu, 04 Nov 1999 11:59:28 +0000 From: "Dr. Patrick Felle" To: nih-image@io.ece.drexel.edu Subject: NIH Image and automated image analysis Message-id: <3821751D.596C4587@ucd.ie> Content-type: multipart/alternative; boundary="------------CEF46AD17707635285885EB5" --------------CEF46AD17707635285885EB5 Content-Type: text/plain; charset=us-ascii; x-mac-type="54455854"; x-mac-creator="4D4F5353" Content-Transfer-Encoding: 7bit I have been using a PC-based imaging system (Optimas running on NT, connected to automated stage light microscope). Will NIH Image run such a system on Macintosh G3 platform, if so what Image Grabber and Automatged stage is recommended? -- ----------------------------------- ----------------------------------- Dr. Patrick Felle (Director) Centre for Healthcare Informatics University College Dublin Earlsfort Terrace Dublin 2 IRELAND Tel / Fax: +353-1-706-7225 Mobile: 086 825 6045 --------------CEF46AD17707635285885EB5 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit I have been using a PC-based imaging system (Optimas running on NT, connected to automated stage light microscope).  Will NIH Image run such a system on Macintosh G3 platform, if so what Image Grabber and Automatged stage is recommended?
--


Dr. Patrick Felle (Director)
Centre for Healthcare Informatics
University College Dublin
Earlsfort Terrace
Dublin 2          IRELAND
Tel / Fax:  +353-1-706-7225     Mobile:  086 825 6045
--------------CEF46AD17707635285885EB5-- ------------------------------ Date: Thu, 4 Nov 1999 14:55:44 +0100 From: Norbert Vischer To: nih-image@io.ece.drexel.edu Subject: Re: "survey mode" in Object Image Message-Id: Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit > There are times when > "survey mode" (command key—space bar) does not display the entire image >in a fit-to-screen mode in Object Image. I suspect that this may be a >function of memory, because it typically happens when I am working on >fairly sizeable files—any thoughts? >Thanks, >Pat Pringle > >-- >"Science is a way of trying not to fool yourself." (Richard Feynman) > "Survey mode" (Command-SpaceBar) should temporarily have the same effect as clicking in the 'fit to screen' box on the right hand side of the title bar, except that a symbolized, movable window frame and a magnification menu become visible. Do you see any further difference? Do you use multiple monitors? Which version do you use? FitToScreen uses nearest neighbor scaling, i.e. small structures may disappear completely when they fall between the pixel grid due to extreme scaling-down. Norbert Vischer University of Amsterdam scientific engineer Molecular Cell Biology Kruislaan 316 tel. +31-20-525-6267(fax 6271) NL-1098 SM Amsterdam e-mail: vischer@bio.uva.nl The Netherlands http://simon.bio.uva.nl/object-image.html ------------------------------ Date: Fri, 5 Nov 1999 16:06:56 +0900 (JST) From: Kardi Teknomo To: nih-image@io.ece.drexel.edu cc: nih-image-d@io.ece.drexel.edu Subject: Perimeter too long Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII Dear Imager, When analyzing particles, I often get message 'perimeter too long' and then its stop. Is anyone can help what is the meaning by 'perimeter too long' and how to solve that problem? Thank you in advance. Kardi -------------------------------- End of nih-image-d Digest V99 Issue #253 **************************************** From nih-image-request@io.ece.drexel.edu Fri Nov 5 02:25 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id CAA24950; Fri, 5 Nov 1999 02:25:40 -0500 (EST) Resent-Date: Fri, 5 Nov 1999 02:25:40 -0500 (EST) Date: Fri, 5 Nov 1999 16:06:56 +0900 (JST) From: Kardi Teknomo X-Sender: kardi@mws To: nih-image@io.ece.drexel.edu cc: nih-image-d@io.ece.drexel.edu Subject: Perimeter too long In-Reply-To: <199911040104.UAA10872@io.ece.drexel.edu> Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"XkHtj2.0.dE5.m8e8u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1919 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 223 Status: O Dear Imager, When analyzing particles, I often get message 'perimeter too long' and then its stop. Is anyone can help what is the meaning by 'perimeter too long' and how to solve that problem? Thank you in advance. Kardi From nih-image-request@io.ece.drexel.edu Fri Nov 5 03:25 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id DAA04739; Fri, 5 Nov 1999 03:25:46 -0500 (EST) Resent-Date: Fri, 5 Nov 1999 03:25:46 -0500 (EST) From: "Leonardo Borghi" To: Subject: nih-image-d Digest Date: Thu, 4 Nov 1999 06:12:17 -0200 Message-ID: <01bf269c$4f678040$0100007f@osiris.ccard.com.br> MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 Resent-Message-ID: <"6cCM41.0.zX.K4f8u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1920 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; boundary="----------------------------"; charset="iso-8859-1" Content-Length: 13 Status: O Unsubscribe From nih-image-request@io.ece.drexel.edu Fri Nov 5 12:41 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA29851; Fri, 5 Nov 1999 12:41:27 -0500 (EST) Resent-Date: Fri, 5 Nov 1999 12:41:27 -0500 (EST) Message-Id: <38231237.57DD866B@maroon.tc.umn.edu> Date: Fri, 05 Nov 1999 11:22:02 -0600 From: "Michael J. Herron" Reply-To: herro001@maroon.tc.umn.edu Organization: U of MN, Deaprtment of Medicine,Infectious Diseases X-Mailer: Mozilla 4.7 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: NIH Image Mailing List Subject: Spot aquisition Content-Transfer-Encoding: 7bit Resent-Message-ID: <"t2S5m.0.jk6.f9n8u"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1921 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 472 Status: O All, I was wondering if it would be possible to aquire images at timed intervals using the Twain that comes with Spot cameras and NIH Image? Thanks, Mike -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ From nih-image-request@io.ece.drexel.edu Fri Nov 5 17:34 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id RAA28778; Fri, 5 Nov 1999 17:34:28 -0500 (EST) Resent-Date: Fri, 5 Nov 1999 17:34:28 -0500 (EST) Message-ID: <382337FA.6C26@mail.smu.edu> Date: Fri, 05 Nov 1999 14:03:09 -0600 From: Duncan Young Organization: Student at SMU X-Mailer: Mozilla 3.0Gold (Macintosh; I; PPC) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #252 References: <199911040116.UAA13521@io.ece.drexel.edu> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"GIepp3.0.OP6.IYr8u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1922 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 344 Status: O Hi there, I am just reporting that Image 1.62 macros for the loading and analysis of the NASA Magellan/Venus Synthetic Apature Rader dataset are now available at -Duncan Young Department of Geological Sciences Southern Methodist University Dallas TX 75275 USA From nih-image-d-request@io.ece.drexel.edu Sat Nov 6 06:13 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA20320; Sat, 6 Nov 1999 06:13:24 -0500 (EST) Date: Sat, 6 Nov 1999 06:13:24 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911061113.GAA20320@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #254 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/254 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 2305 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 254 Today's Topics: nih-image-d Digest [ "Leonardo Borghi" To: Subject: nih-image-d Digest Message-ID: <01bf269c$4f678040$0100007f@osiris.ccard.com.br> Content-Transfer-Encoding: 7bit Content-Type: text/plain; boundary="----------------------------"; charset="iso-8859-1" Unsubscribe ------------------------------ Date: Fri, 05 Nov 1999 11:22:02 -0600 From: "Michael J. Herron" To: NIH Image Mailing List Subject: Spot aquisition Message-Id: <38231237.57DD866B@maroon.tc.umn.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit All, I was wondering if it would be possible to aquire images at timed intervals using the Twain that comes with Spot cameras and NIH Image? Thanks, Mike -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ ------------------------------ Date: Fri, 05 Nov 1999 14:03:09 -0600 From: Duncan Young To: nih-image@io.ece.drexel.edu Subject: Re: nih-image-d Digest V99 #252 Message-ID: <382337FA.6C26@mail.smu.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi there, I am just reporting that Image 1.62 macros for the loading and analysis of the NASA Magellan/Venus Synthetic Apature Rader dataset are now available at -Duncan Young Department of Geological Sciences Southern Methodist University Dallas TX 75275 USA -------------------------------- End of nih-image-d Digest V99 Issue #254 **************************************** From nih-image-request@io.ece.drexel.edu Sat Nov 6 12:04 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA26233; Sat, 6 Nov 1999 12:04:44 -0500 (EST) Resent-Date: Sat, 6 Nov 1999 12:04:44 -0500 (EST) Mime-Version: 1.0 X-Sender: lis3@pop.lis.ch Message-Id: In-Reply-To: <002f01bf264b$27839240$0200a8c0@internet> References: <002f01bf264b$27839240$0200a8c0@internet> Date: Sat, 6 Nov 1999 16:46:06 +0100 To: nih-image@io.ece.drexel.edu From: Beat Ludin Subject: Re: Contrast and Brightness Resent-Message-ID: <"Xdd8o3.0.pi5.bm59u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1923 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" ; format="flowed" Content-Length: 4702 Status: O >Hi all, > >Ever have one of those questions that you thought was obvious until you >tried to answer it in detail. The question I had was "How does NIH Image >change the palette in response to the brightness and contrast controls". >Sounds obvious, but then some of the following issues came up. > >*** As a starting assumption for gray scale images *** >Brightness implies that a constant is added to the baseline grayscale >palette values. >Contrast implies a change in slope, i.e. the baseline palette values are >multiplied by the contrast slope. >The modified gray palette values should be -- new = slope * original + >constant -- where the original values for a gray scale run from 0-255, the >slope is the contrast control (values from zero to infinity) and the >constant is the brightness control (values from -255 to +255). > >This is the obvious part. Here are some of the sticky details. > >1) Although I am reasonably certain that the NIH Image brightness and >contrast controls operate as indicated above, Not quite. In your function (output=contrast*input+offset), the contrast hinges on a pixel value of 0%, i.e. an output of 0 remains constant regardless of contrast. In NI's function, the contrast hinges on 50% of the dynamic range, so the function becomes something like output=contrast*(input+offset-127)+127. Note that I'm using 'offest' instead of 'brightness', the reason for that is the answer to 2) > I have been shown other >programming texts that use different formulas for affecting contrast and >brightness. Is there a standard definition of contrast and brightness in >digital image processing? I don't think so. Especially when it comes to the choice of the contrast hinge point. > >2) As the contrast control is changed, the brightness level in NIH Image is >seen to change slightly. The above assumptions would indicate that the two >parameters should be independent. Why is this not the case? You may have noticed that this is only the case when output clipping occurs, i.e. the function an input of 0 produces an output > 0 or and/or an input of 255 produces and output > 255. As long as there is no clipping, offset and brightness values are identical. NI's brightness hinges on 50% of the output range (e.g. 34-192) and and not 50% of the dynamic range (0-255). When you change the slope, NI keeps the offset constant, which means that the brightness changes. I guess this is due to the graphical representation in NI's Map window, where the middle hindle represents the brightness. Confused? It might become more obvious when you play around with the Map window (note how an output of 50% (y=127 remains constant when you change the contrast). >3) In a grayscale image all the red, green, blue (RGB) values are set equal. >Thus, the formula given above is simply applied to each of the separate >color values. When a false color palette is used, the mapping between the >above formula and the RGB values is rather unclear. NIH Image does what I >would intuitively expect, but how does it actually handling this mapping? This becomes obvious with the concept of lookup tables (LUT, palette). Pseudocolor images (also called 'indexed color' images) are handled like grayscale images, but every output value is mapped to an color value using a LUT. This is a table that assigns an arbitrary RGB value to every output value. >4) The point of theses details has to do with opening a TIFF file saved from >NIH Image in another application. Somehow, when NIH Image opens an image >that it created, it is able to restore the brightness and contrast settings >for that image - even with false color palettes. However, the standard TIFF >file format only contains a palette - brightness and contrast are not part >of standard. Thus, other applications with brightness and contrast controls >do not recover the values set in NIH Image. How, exactly does NIH Image >recover the brightness and contrast values, especially if a false color >palette is being used? Are these values being stored somewhere in the TIFF >file? LUT, contrast and brightness settings are saved in the header of indexed color images. Let me conclude by saying that all of this deduced from empirical observations. I'm sure, Wayne could give you a more concise and precise explanation. Cheers, Beat LIFE IMAGING SERVICES - visit our web site at http://www.lis.ch +----------------------------------------------------------- | Dr. B. Ludin | Life Imaging Services fon ++41 (0)79 235 7154 | Muehletalweg 22 fax ++41 (0)86 062 296 3160 NEW! | CH-4600 Olten beat.ludin@lis.ch | Switzerland http://www.lis.ch From nih-image-d-request@io.ece.drexel.edu Sun Nov 7 06:16 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA15433; Sun, 7 Nov 1999 06:16:58 -0500 (EST) Date: Sun, 7 Nov 1999 06:16:58 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911071116.GAA15433@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #255 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/255 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5295 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 255 Today's Topics: Re: Contrast and Brightness [ Beat Ludin ] ------------------------------ Date: Sat, 6 Nov 1999 16:46:06 +0100 From: Beat Ludin To: nih-image@io.ece.drexel.edu Subject: Re: Contrast and Brightness Message-Id: Content-Type: text/plain; charset="us-ascii" ; format="flowed" >Hi all, > >Ever have one of those questions that you thought was obvious until you >tried to answer it in detail. The question I had was "How does NIH Image >change the palette in response to the brightness and contrast controls". >Sounds obvious, but then some of the following issues came up. > >*** As a starting assumption for gray scale images *** >Brightness implies that a constant is added to the baseline grayscale >palette values. >Contrast implies a change in slope, i.e. the baseline palette values are >multiplied by the contrast slope. >The modified gray palette values should be -- new = slope * original + >constant -- where the original values for a gray scale run from 0-255, the >slope is the contrast control (values from zero to infinity) and the >constant is the brightness control (values from -255 to +255). > >This is the obvious part. Here are some of the sticky details. > >1) Although I am reasonably certain that the NIH Image brightness and >contrast controls operate as indicated above, Not quite. In your function (output=contrast*input+offset), the contrast hinges on a pixel value of 0%, i.e. an output of 0 remains constant regardless of contrast. In NI's function, the contrast hinges on 50% of the dynamic range, so the function becomes something like output=contrast*(input+offset-127)+127. Note that I'm using 'offest' instead of 'brightness', the reason for that is the answer to 2) > I have been shown other >programming texts that use different formulas for affecting contrast and >brightness. Is there a standard definition of contrast and brightness in >digital image processing? I don't think so. Especially when it comes to the choice of the contrast hinge point. > >2) As the contrast control is changed, the brightness level in NIH Image is >seen to change slightly. The above assumptions would indicate that the two >parameters should be independent. Why is this not the case? You may have noticed that this is only the case when output clipping occurs, i.e. the function an input of 0 produces an output > 0 or and/or an input of 255 produces and output > 255. As long as there is no clipping, offset and brightness values are identical. NI's brightness hinges on 50% of the output range (e.g. 34-192) and and not 50% of the dynamic range (0-255). When you change the slope, NI keeps the offset constant, which means that the brightness changes. I guess this is due to the graphical representation in NI's Map window, where the middle hindle represents the brightness. Confused? It might become more obvious when you play around with the Map window (note how an output of 50% (y=127 remains constant when you change the contrast). >3) In a grayscale image all the red, green, blue (RGB) values are set equal. >Thus, the formula given above is simply applied to each of the separate >color values. When a false color palette is used, the mapping between the >above formula and the RGB values is rather unclear. NIH Image does what I >would intuitively expect, but how does it actually handling this mapping? This becomes obvious with the concept of lookup tables (LUT, palette). Pseudocolor images (also called 'indexed color' images) are handled like grayscale images, but every output value is mapped to an color value using a LUT. This is a table that assigns an arbitrary RGB value to every output value. >4) The point of theses details has to do with opening a TIFF file saved from >NIH Image in another application. Somehow, when NIH Image opens an image >that it created, it is able to restore the brightness and contrast settings >for that image - even with false color palettes. However, the standard TIFF >file format only contains a palette - brightness and contrast are not part >of standard. Thus, other applications with brightness and contrast controls >do not recover the values set in NIH Image. How, exactly does NIH Image >recover the brightness and contrast values, especially if a false color >palette is being used? Are these values being stored somewhere in the TIFF >file? LUT, contrast and brightness settings are saved in the header of indexed color images. Let me conclude by saying that all of this deduced from empirical observations. I'm sure, Wayne could give you a more concise and precise explanation. Cheers, Beat LIFE IMAGING SERVICES - visit our web site at http://www.lis.ch +----------------------------------------------------------- | Dr. B. Ludin | Life Imaging Services fon ++41 (0)79 235 7154 | Muehletalweg 22 fax ++41 (0)86 062 296 3160 NEW! | CH-4600 Olten beat.ludin@lis.ch | Switzerland http://www.lis.ch -------------------------------- End of nih-image-d Digest V99 Issue #255 **************************************** From nih-image-request@io.ece.drexel.edu Sun Nov 7 08:22 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id IAA08597; Sun, 7 Nov 1999 08:22:28 -0500 (EST) Resent-Date: Sun, 7 Nov 1999 08:22:28 -0500 (EST) From: "Leonardo Borghi" To: Subject: Unsubscribe Date: Sat, 6 Nov 1999 11:12:55 -0200 Message-ID: <01bf2858$a3dd84e0$8e95f4c8@osiris.ccard.com.br> MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 4.71.1712.3 X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3 Resent-Message-ID: <"IDzER2.0.wZ1.7gN9u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1924 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; boundary="----------------------------"; charset="iso-8859-1" Content-Length: 13 Status: O Unsubscribe From nih-image-request@io.ece.drexel.edu Sun Nov 7 13:18 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id NAA00569; Sun, 7 Nov 1999 13:18:31 -0500 (EST) Resent-Date: Sun, 7 Nov 1999 13:18:31 -0500 (EST) X-Sender: francus@eclogite.geo.umass.edu Message-Id: Mime-Version: 1.0 X-Mailer: Eudora F1.5.3 Date: Sun, 7 Nov 1999 13:05:02 -0500 To: nih-image@io.ece.drexel.edu From: francus@geo.umass.edu (Pierre Francus) Subject: SEM image acquisition Resent-Message-ID: <"VU4FW2.0.8o6.byR9u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1925 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 825 Status: O Dear Imagers, We intend to purchase a frame grabber to acquire images from a SEM (JEOL 5410). We would like to acquire images of at least 1024 x 1024 pixels. To obtain such images, is the video output of the SEM good enough ? Can we simply plug the frame grabber in that output or do we have to build some fancy system to grab images elswhere in the SEM ? Since we need digital images urgently, we are looking for a "Plug & PLay" system. Thanks in advance, Pierre ***************************************** Pierre FRANCUS, Ph.D. PostDoctoral Research Associate University of Massachusetts Department of Geosciences Morrill Science Center Amherst - MA 01003-5820 Phone: 1-413-545-0659 fax: 1-413-545-1200 E-mail: francus@geo.umass.edu http://www.geo.umass.edu/climate/francus/ ***************************************** From nih-image-request@io.ece.drexel.edu Sun Nov 7 19:26 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA03090; Sun, 7 Nov 1999 19:26:18 -0500 (EST) Resent-Date: Sun, 7 Nov 1999 19:26:18 -0500 (EST) Message-ID: <38260FCF.7830@mail.smu.edu> Date: Sun, 07 Nov 1999 17:48:32 -0600 From: Duncan Young Organization: Student at SMU X-Mailer: Mozilla 3.0Gold (Macintosh; I; PPC) MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Magellan Image Macros Content-Transfer-Encoding: 7bit Resent-Message-ID: <"c5rkX.0.I2.tKX9u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1926 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 196 Status: O The Magellan SAR macros at are now Netscape Navigator friendly...thanks to some good spotting by Bill Christens-Barry! -DUNCAN YOUNG From nih-image-d-request@io.ece.drexel.edu Mon Nov 8 05:07 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA17290; Mon, 8 Nov 1999 05:07:47 -0500 (EST) Date: Mon, 8 Nov 1999 05:07:47 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911081007.FAA17290@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #256 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/256 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 3505 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 256 Today's Topics: Unsubscribe [ "Leonardo Borghi" To: Subject: Unsubscribe Message-ID: <01bf2858$a3dd84e0$8e95f4c8@osiris.ccard.com.br> Content-Transfer-Encoding: 7bit Content-Type: text/plain; boundary="----------------------------"; charset="iso-8859-1" Unsubscribe ------------------------------ Date: Sun, 7 Nov 1999 13:05:02 -0500 From: francus@geo.umass.edu (Pierre Francus) To: nih-image@io.ece.drexel.edu Subject: SEM image acquisition Message-Id: Content-Type: text/plain; charset="us-ascii" Dear Imagers, We intend to purchase a frame grabber to acquire images from a SEM (JEOL 5410). We would like to acquire images of at least 1024 x 1024 pixels. To obtain such images, is the video output of the SEM good enough ? Can we simply plug the frame grabber in that output or do we have to build some fancy system to grab images elswhere in the SEM ? Since we need digital images urgently, we are looking for a "Plug & PLay" system. Thanks in advance, Pierre ***************************************** Pierre FRANCUS, Ph.D. PostDoctoral Research Associate University of Massachusetts Department of Geosciences Morrill Science Center Amherst - MA 01003-5820 Phone: 1-413-545-0659 fax: 1-413-545-1200 E-mail: francus@geo.umass.edu http://www.geo.umass.edu/climate/francus/ ***************************************** ------------------------------ Date: Sun, 07 Nov 1999 17:48:32 -0600 From: Duncan Young To: nih-image@io.ece.drexel.edu Subject: Magellan Image Macros Message-ID: <38260FCF.7830@mail.smu.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit The Magellan SAR macros at are now Netscape Navigator friendly...thanks to some good spotting by Bill Christens-Barry! -DUNCAN YOUNG ------------------------------ Date: Mon, 08 Nov 1999 11:48:48 +0200 From: "Dr. Ron Goldstein" To: nih-image@io.ece.drexel.edu Subject: Kodak MDS 100 Message-ID: <38269C80.F6528556@mail.biu.ac.il> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, Has anyone tried the new Kodak MDS 100 microscope digital camera? It give megapixel files, but uses interpolation to do so, so I am concerned about the image quality. Any information would be appreciated, off list is fine. Thanks, -Ron Goldstein -- ----------------------------------------------------------------------------- Dr. Ron Goldstein email goldst@mail.biu.ac.il Faculty Life Sciences Tel. 972-3-531-8216 Bar-Ilan Univ FAX 972-3-535-1824 52900 Ramat-Gan, ISRAEL http://www.biu.ac.il/LS/staff/goldstein.html ----------------------------------------------------------------------------- -------------------------------- End of nih-image-d Digest V99 Issue #256 **************************************** From nih-image-request@io.ece.drexel.edu Mon Nov 8 05:13 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA18467; Mon, 8 Nov 1999 05:13:34 -0500 (EST) Resent-Date: Mon, 8 Nov 1999 05:13:34 -0500 (EST) Message-ID: <38269C80.F6528556@mail.biu.ac.il> Date: Mon, 08 Nov 1999 11:48:48 +0200 From: "Dr. Ron Goldstein" X-Mailer: Mozilla 4.7 [en] (Win98; I) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Kodak MDS 100 Content-Transfer-Encoding: 7bit Resent-Message-ID: <"FNiuM1.0.AR3.-qf9u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1927 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 694 Status: O Hi, Has anyone tried the new Kodak MDS 100 microscope digital camera? It give megapixel files, but uses interpolation to do so, so I am concerned about the image quality. Any information would be appreciated, off list is fine. Thanks, -Ron Goldstein -- ----------------------------------------------------------------------------- Dr. Ron Goldstein email goldst@mail.biu.ac.il Faculty Life Sciences Tel. 972-3-531-8216 Bar-Ilan Univ FAX 972-3-535-1824 52900 Ramat-Gan, ISRAEL http://www.biu.ac.il/LS/staff/goldstein.html ----------------------------------------------------------------------------- From nih-image-request@io.ece.drexel.edu Mon Nov 8 06:30 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA02781; Mon, 8 Nov 1999 06:30:26 -0500 (EST) Resent-Date: Mon, 8 Nov 1999 06:30:26 -0500 (EST) Message-ID: From: Gary Chinga To: "'nih-image@io.ece.drexel.edu'" Subject: RE: Magellan Image Macros Date: Mon, 8 Nov 1999 12:16:22 +0100 X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.996.62 MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Resent-Message-ID: <"F0jdB.0.qB7.O1h9u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1928 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 515 Status: O It seems that your macros are not completed in the web page, the last part of the first and second set of macros is missing. Gary. >-----Original Message----- >From: Duncan Young [SMTP:dyoung@post.cis.smu.edu] >Sent: 8. november 1999 00:49 >To: nih-image@io.ece.drexel.edu >Subject: Magellan Image Macros > >The Magellan SAR macros at > are now Netscape >Navigator friendly...thanks to some good spotting by Bill >Christens-Barry! > -DUNCAN YOUNG > From nih-image-request@io.ece.drexel.edu Mon Nov 8 07:05 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA09795; Mon, 8 Nov 1999 07:05:06 -0500 (EST) Resent-Date: Mon, 8 Nov 1999 07:05:06 -0500 (EST) X-Sender: fzhao@pop.helsinki.fi Message-Id: In-Reply-To: Mime-Version: 1.0 Date: Mon, 8 Nov 1999 13:47:39 +0300 To: nih-image@io.ece.drexel.edu From: Fang Zhao Subject: Re: SEM image acquisition Resent-Message-ID: <"xDeRZ2.0.yQ1.bVh9u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1929 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 1342 Status: O >Dear Imagers, > >We intend to purchase a frame grabber to acquire images from a SEM (JEOL >5410). We would like to acquire images of at least 1024 x 1024 pixels. To >obtain such images, is the video output of the SEM good enough ? Can we >simply plug the frame grabber in that output or do we have to build some >fancy system to grab images elswhere in the SEM ? > Yes, the video output from the SEM must good enough for your purpose. According to my knowledge, in all JEOL SEM, the same signal is directed to the photographic film through the analogue slow scanning system which gives, depending on microscope and recording time, is typically 1900 x 1400 pixels. JEOL's SemAfore video capture card detect the signal via an existing socket on SEM and make analogue to digital converting, which has the same Res as that after the slow scaning. I have expirence on our old JSM-820 equiped with SemAfore, which is a ISA card and claimed 12 bit image resolution. The software unfortunately, is PC based: win31. 95/98, NT. The images I acquired are up to 1652 X 1226. Visit http://www.jeol.se for more information. Sorry,I think there must be a web site in US, but I don't know it. Fang Zhao, M.D., Ph.D. Department of Pathology Haartman Institute P.O box 00014 University of Helsinki Tel +358 9 1912 6261 email: fang.zhao@helsinki.fi From nih-image-request@io.ece.drexel.edu Mon Nov 8 12:51 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA13732; Mon, 8 Nov 1999 12:51:21 -0500 (EST) Resent-Date: Mon, 8 Nov 1999 12:51:21 -0500 (EST) Message-Id: In-Reply-To: <199911080952.EAA14073@io.ece.drexel.edu> Mime-Version: 1.0 Date: Mon, 8 Nov 1999 12:33:15 -0500 To: nih-image@io.ece.drexel.edu From: Bala Iyengar Subject: macbrick macio module Cc: iyengabg@mcmaster.ca Resent-Message-ID: <"UGQDW2.0.sn2.Zbm9u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1930 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 244 Status: O Hello all, Is there anyone out there who has used the MacIO module made by Macbrick (http://www.w4r.com/macbrick/macio.html) to control serial devices? any pointers as to how to control this using Image will be greatly appreciated! TIA. Bala From nih-image-request@io.ece.drexel.edu Mon Nov 8 14:03 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id OAA23171; Mon, 8 Nov 1999 14:03:16 -0500 (EST) Resent-Date: Mon, 8 Nov 1999 14:03:16 -0500 (EST) Message-Id: In-Reply-To: References: <199911040104.UAA10872@io.ece.drexel.edu> Mime-Version: 1.0 Date: Mon, 8 Nov 1999 14:54:20 -0400 To: nih-image@io.ece.drexel.edu From: Wayne Rasband Subject: Re: Perimeter too long Resent-Message-ID: <"pqztV3.0.k05.ldn9u"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1931 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 844 Status: O >Dear Imager, >When analyzing particles, I often get message 'perimeter too long' and >then its stop. Is anyone can help what is the meaning by 'perimeter too >long' and how to solve that problem? >Thank you in advance. Are you using Scion Image for Windows? It has a bug that generates the 'perimeter too long' message. In NIH Image, this message is displayed when you exceed about 10,000 coordinate pairs in an object outline. You may want to try ImageJ, which has no arbitrary limit on perimeter length. Download version 1.09 from http://rsb.info.nih.gov/ij/ and then upgrade to v1.10 by replacing the ij.jar and ij.properties files in the ImageJ folder with the ones at ftp://codon.nih.gov/pub/image-j/ Thresholding, particle analysis and presentation of results are improved in v1.10. See the release notes for details. -wayne From nih-image-d-request@io.ece.drexel.edu Tue Nov 9 06:12 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA13237; Tue, 9 Nov 1999 06:12:12 -0500 (EST) Date: Tue, 9 Nov 1999 06:12:12 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911091112.GAA13237@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #257 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/257 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 6006 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 257 Today's Topics: RE: Magellan Image Macros [ Gary Chinga ] Re: SEM image acquisition [ Fang Zhao ] macbrick macio module [ Bala Iyengar ] can't do live paste [ Gary Radice ] ------------------------------ Date: Mon, 8 Nov 1999 12:16:22 +0100 From: Gary Chinga To: "'nih-image@io.ece.drexel.edu'" Subject: RE: Magellan Image Macros Message-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit It seems that your macros are not completed in the web page, the last part of the first and second set of macros is missing. Gary. >-----Original Message----- >From: Duncan Young [SMTP:dyoung@post.cis.smu.edu] >Sent: 8. november 1999 00:49 >To: nih-image@io.ece.drexel.edu >Subject: Magellan Image Macros > >The Magellan SAR macros at > are now Netscape >Navigator friendly...thanks to some good spotting by Bill >Christens-Barry! > -DUNCAN YOUNG > ------------------------------ Date: Mon, 8 Nov 1999 13:47:39 +0300 From: Fang Zhao To: nih-image@io.ece.drexel.edu Subject: Re: SEM image acquisition Message-Id: Content-Type: text/plain; charset="us-ascii" >Dear Imagers, > >We intend to purchase a frame grabber to acquire images from a SEM (JEOL >5410). We would like to acquire images of at least 1024 x 1024 pixels. To >obtain such images, is the video output of the SEM good enough ? Can we >simply plug the frame grabber in that output or do we have to build some >fancy system to grab images elswhere in the SEM ? > Yes, the video output from the SEM must good enough for your purpose. According to my knowledge, in all JEOL SEM, the same signal is directed to the photographic film through the analogue slow scanning system which gives, depending on microscope and recording time, is typically 1900 x 1400 pixels. JEOL's SemAfore video capture card detect the signal via an existing socket on SEM and make analogue to digital converting, which has the same Res as that after the slow scaning. I have expirence on our old JSM-820 equiped with SemAfore, which is a ISA card and claimed 12 bit image resolution. The software unfortunately, is PC based: win31. 95/98, NT. The images I acquired are up to 1652 X 1226. Visit http://www.jeol.se for more information. Sorry,I think there must be a web site in US, but I don't know it. Fang Zhao, M.D., Ph.D. Department of Pathology Haartman Institute P.O box 00014 University of Helsinki Tel +358 9 1912 6261 email: fang.zhao@helsinki.fi ------------------------------ Date: Mon, 8 Nov 1999 12:33:15 -0500 From: Bala Iyengar To: nih-image@io.ece.drexel.edu Cc: iyengabg@mcmaster.ca Subject: macbrick macio module Message-Id: Content-Type: text/plain; charset="us-ascii" Hello all, Is there anyone out there who has used the MacIO module made by Macbrick (http://www.w4r.com/macbrick/macio.html) to control serial devices? any pointers as to how to control this using Image will be greatly appreciated! TIA. Bala ------------------------------ Date: Mon, 8 Nov 1999 14:54:20 -0400 From: Wayne Rasband To: nih-image@io.ece.drexel.edu Subject: Re: Perimeter too long Message-Id: Content-Type: text/plain; charset="us-ascii" >Dear Imager, >When analyzing particles, I often get message 'perimeter too long' and >then its stop. Is anyone can help what is the meaning by 'perimeter too >long' and how to solve that problem? >Thank you in advance. Are you using Scion Image for Windows? It has a bug that generates the 'perimeter too long' message. In NIH Image, this message is displayed when you exceed about 10,000 coordinate pairs in an object outline. You may want to try ImageJ, which has no arbitrary limit on perimeter length. Download version 1.09 from http://rsb.info.nih.gov/ij/ and then upgrade to v1.10 by replacing the ij.jar and ij.properties files in the ImageJ folder with the ones at ftp://codon.nih.gov/pub/image-j/ Thresholding, particle analysis and presentation of results are improved in v1.10. See the release notes for details. -wayne ------------------------------ Date: Mon, 8 Nov 1999 15:54:54 -0500 From: Gary Radice To: nih-image-d@io.ece.drexel.edu Subject: can't do live paste Message-Id: Content-Type: text/plain; charset="us-ascii" I have just run into an odd problem. I am no longer able to do a live paste. I'm trying to register serial sections for a 3D reconstruction. With a previously saved image window open, I grab a second live image. Then I do a "select all." When I "select all" the title bar goes from "camera (live) to "camera," and it is no longer a live image. When I try to paste onto the first image, with the paste control window open and "live paste" selected, it simply pastes the static second image. Happens with NIH Image 1.62b, Object Image, and Scion Image on two different machines, both running Mac OS 8.1. I've done this procedure a thousand times without a problem. This is probably something simple but I can't figure it out. What am I forgetting? Gary P. Radice gradice@richmond.edu Associate Professor of Biology 804 289 8107 (voice) University of Richmond 804 289 8233 (FAX) Richmond VA 23173 http://www.science.richmond.edu/~radice -------------------------------- End of nih-image-d Digest V99 Issue #257 **************************************** From nih-image-request@io.ece.drexel.edu Tue Nov 9 21:48 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id VAA10743; Tue, 9 Nov 1999 21:48:57 -0500 (EST) Resent-Date: Tue, 9 Nov 1999 21:48:57 -0500 (EST) X-Authentication-Warning: sun-2.cae.wisc.edu: yinl owned process doing -bs Date: Tue, 9 Nov 1999 20:31:27 -0600 (CST) From: Lei Yin To: nih-image@io.ece.drexel.edu Subject: crop Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"NdG3c3.0.Jj1.3aDAu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1932 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 437 Status: O Hi, I have acquired a movie of TIFF stored in stack. However, I only need one fixed portion of the image to analyze for the area image by image. Do you have any suggestions of how to do this? Does NIH-Image has a crop tool which can apply to the stack? Or I need to write a special Macro to crop the rectangular are I need for analysis? Also, I wonder if you have any method for meausring area of dendrites? I appreciate your help. Lei From nih-image-request@io.ece.drexel.edu Tue Nov 9 22:29 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id WAA19429; Tue, 9 Nov 1999 22:29:51 -0500 (EST) Resent-Date: Tue, 9 Nov 1999 22:29:51 -0500 (EST) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: yinl@cae.wisc.edu, nih-image@io.ece.drexel.edu Date: Wed, 10 Nov 1999 14:23:54 +1100 Subject: Re: crop Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: Resent-Message-ID: <"u3-zc3.0.084.MGEAu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1933 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 852 Status: O >Date: Tue, 9 Nov 1999 20:31:27 -0600 (CST) >From: Lei Yin >To: nih-image@io.ece.drexel.edu >Subject: crop > >Hi, >I have acquired a movie of TIFF stored in stack. >However, I only need one fixed portion of the image to analyze for the >area image by image. Do you have any suggestions of how to do this? >Does >NIH-Image has a crop tool which can apply to the stack? Or I need to >write >a special Macro to crop the rectangular are I need for analysis? >Also, I wonder if you have any method for meausring area of dendrites? >I appreciate your help. >Lei > see Crop and Scale macro (Stacks macros (macros which accompany NIH-Image at website Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Tue Nov 9 22:42 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id WAA22275; Tue, 9 Nov 1999 22:42:34 -0500 (EST) Resent-Date: Tue, 9 Nov 1999 22:42:34 -0500 (EST) Message-Id: <3.0.5.32.19991109223846.00a903b0@codon.nih.gov> X-Sender: wayne@codon.nih.gov X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Tue, 09 Nov 1999 22:38:46 -0500 To: nih-image@io.ece.drexel.edu From: Wayne Rasband Subject: Re: crop In-Reply-To: Mime-Version: 1.0 Resent-Message-ID: <"b9kFo1.0.Em4.rQEAu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1934 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 591 Status: O At 08:31 PM 11/9/99 -0600, you wrote: >Hi, >I have acquired a movie of TIFF stored in stack. >However, I only need one fixed portion of the image to analyze for the >area image by image. Do you have any suggestions of how to do this? Does >NIH-Image has a crop tool which can apply to the stack? Or I need to write >a special Macro to crop the rectangular are I need for analysis? >Also, I wonder if you have any method for meausring area of dendrites? The "Crop and Scale-fast" macro in the "Stacks" macro file will crop a stack. Or you can use the Image/Crop command in ImageJ. -wayne From nih-image-d-request@io.ece.drexel.edu Thu Nov 11 06:15 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA16890; Thu, 11 Nov 1999 06:15:37 -0500 (EST) Date: Thu, 11 Nov 1999 06:15:37 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911111115.GAA16890@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #258 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/258 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 3105 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 258 Today's Topics: crop [ Lei Yin ] Re: crop [ GJOSS@rna.bio.mq.edu.au ] Re: crop [ Wayne Rasband ] ------------------------------ Date: Tue, 9 Nov 1999 20:31:27 -0600 (CST) From: Lei Yin To: nih-image@io.ece.drexel.edu Subject: crop Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, I have acquired a movie of TIFF stored in stack. However, I only need one fixed portion of the image to analyze for the area image by image. Do you have any suggestions of how to do this? Does NIH-Image has a crop tool which can apply to the stack? Or I need to write a special Macro to crop the rectangular are I need for analysis? Also, I wonder if you have any method for meausring area of dendrites? I appreciate your help. Lei ------------------------------ Date: Wed, 10 Nov 1999 14:23:54 +1100 From: GJOSS@rna.bio.mq.edu.au To: yinl@cae.wisc.edu, nih-image@io.ece.drexel.edu Subject: Re: crop Message-ID: >Date: Tue, 9 Nov 1999 20:31:27 -0600 (CST) >From: Lei Yin >To: nih-image@io.ece.drexel.edu >Subject: crop > >Hi, >I have acquired a movie of TIFF stored in stack. >However, I only need one fixed portion of the image to analyze for the >area image by image. Do you have any suggestions of how to do this? >Does >NIH-Image has a crop tool which can apply to the stack? Or I need to >write >a special Macro to crop the rectangular are I need for analysis? >Also, I wonder if you have any method for meausring area of dendrites? >I appreciate your help. >Lei > see Crop and Scale macro (Stacks macros (macros which accompany NIH-Image at website Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Tue, 09 Nov 1999 22:38:46 -0500 From: Wayne Rasband To: nih-image@io.ece.drexel.edu Subject: Re: crop Message-Id: <3.0.5.32.19991109223846.00a903b0@codon.nih.gov> Content-Type: text/plain; charset="us-ascii" At 08:31 PM 11/9/99 -0600, you wrote: >Hi, >I have acquired a movie of TIFF stored in stack. >However, I only need one fixed portion of the image to analyze for the >area image by image. Do you have any suggestions of how to do this? Does >NIH-Image has a crop tool which can apply to the stack? Or I need to write >a special Macro to crop the rectangular are I need for analysis? >Also, I wonder if you have any method for meausring area of dendrites? The "Crop and Scale-fast" macro in the "Stacks" macro file will crop a stack. Or you can use the Image/Crop command in ImageJ. -wayne -------------------------------- End of nih-image-d Digest V99 Issue #258 **************************************** From nih-image-request@io.ece.drexel.edu Thu Nov 11 12:10 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id MAA24887; Thu, 11 Nov 1999 12:10:29 -0500 (EST) Resent-Date: Thu, 11 Nov 1999 12:10:29 -0500 (EST) Mime-Version: 1.0 Message-Id: Date: Thu, 11 Nov 1999 10:48:26 -0600 To: nih-image@io.ece.drexel.edu From: Dorothy Schafer Subject: fourier analysis on a movie Cc: wayne@codon.nih.gov Resent-Message-ID: <"_RF1-2.0.vz4.aDlAu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1935 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" ; format="flowed" Content-Length: 947 Status: O Hi, We are interested in determining a measure of the fluxuation in fluorescence intensity over time, and in comparing that flux measurement on a cell to cell basis, so that data from different movies can be compared. A fourier transform analysis of the fluorescence intensity over time would do this. Within NIHImage, the standard FFT macros perform a two-dimensional fourier transform on a region of a single square image. Is there a macro that performs a three-dimensional fourier transform on a stack corresponding to a single region over time (as in a movie)? Thanks, Dorothy Schafer Dorothy Schafer Dept. of Cell Biology-Box 8228 Washington University School of Medicine 660 S. Euclid Ave. St. Louis, MO 63110 Ph: 314-362-0098 FAX: 314-362-0098 World Wide Web site: http://www.cooperlab.wustl.edu/ for FedEx/UPS Shipping: Dept. of Cell Biology Washington University School of Medicine 4566 Scott Ave. St. Louis, MO 63110 From nih-image-request@io.ece.drexel.edu Thu Nov 11 16:30 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA15477; Thu, 11 Nov 1999 16:30:16 -0500 (EST) Resent-Date: Thu, 11 Nov 1999 16:30:16 -0500 (EST) X-Sender: sbudick1@swarthmore.edu (Unverified) Message-Id: In-Reply-To: <3.0.5.32.19991109223846.00a903b0@codon.nih.gov> References: Mime-Version: 1.0 Date: Thu, 11 Nov 1999 16:10:54 -0400 To: nih-image@io.ece.drexel.edu From: Seth Budick Subject: Re: crop Resent-Message-ID: <"jf7Nn1.0.kx2.Y5pAu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1936 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 915 Status: O >At 08:31 PM 11/9/99 -0600, you wrote: >>Hi, >>I have acquired a movie of TIFF stored in stack. >>However, I only need one fixed portion of the image to analyze for the >>area image by image. Do you have any suggestions of how to do this? Does >>NIH-Image has a crop tool which can apply to the stack? Or I need to write >>a special Macro to crop the rectangular are I need for analysis? >>Also, I wonder if you have any method for meausring area of dendrites? > >The "Crop and Scale-fast" macro in the "Stacks" macro file will crop a >stack. Or you can use the Image/Crop command in ImageJ. > >-wayne Is there a way to do this for oval ROIs since Scale and Rotate requires rectangular ROIs? Thanks, Seth ________________________________ Seth Budick Biology Department 414 Mugar Hall Northeastern University Boston, MA 02115 Phone: (617) 373-4034 e-mail: sbudick1@swarthmore.edu http://www.omalleylab.neu.edu From nih-image-request@io.ece.drexel.edu Thu Nov 11 18:04 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA05302; Thu, 11 Nov 1999 18:04:14 -0500 (EST) Resent-Date: Thu, 11 Nov 1999 18:04:14 -0500 (EST) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: sbudick1@swarthmore.edu, nih-image@io.ece.drexel.edu Date: Fri, 12 Nov 1999 9:55:26 +1100 Subject: Re: crop Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <128AB0A4488@rna.bio.mq.edu.au> Resent-Message-ID: <"ozA8S3.0.ca.kWqAu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1937 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 1947 Status: O >Date: Thu, 11 Nov 1999 16:10:54 -0400 >To: nih-image@io.ece.drexel.edu >From: Seth Budick >Subject: Re: crop > >>At 08:31 PM 11/9/99 -0600, you wrote: >>>Hi, >>>I have acquired a movie of TIFF stored in stack. >>>However, I only need one fixed portion of the image to analyze for >the >>>area image by image. Do you have any suggestions of how to do this? >Does >>>NIH-Image has a crop tool which can apply to the stack? Or I need to >write >>>a special Macro to crop the rectangular are I need for analysis? >>>Also, I wonder if you have any method for meausring area of >dendrites? >> >>The "Crop and Scale-fast" macro in the "Stacks" macro file will crop a >>stack. Or you can use the Image/Crop command in ImageJ. >> >>-wayne > >Is there a way to do this for oval ROIs since Scale and Rotate requires >rectangular ROIs? >Thanks, >Seth > >________________________________ >Seth Budick >Biology Department >414 Mugar Hall >Northeastern University >Boston, MA 02115 > >Phone: (617) 373-4034 >e-mail: sbudick1@swarthmore.edu >http://www.omalleylab.neu.edu > 'Croping a movie' normally implies that you wish to produce a smaller frame movie as the output. All movies, stacks, windows, frames are rectangular. You can make oval ROIs within the frame of a window or stack. If made on a stack, it is retained on the stack as you turn to each frame to process through the stack (by selectSlice?) unless the process is one that removes the ROI. In this case you need to reinstate the ROI by any of a number of techniques including use of "restoreRoi","save as(outline" etc. getRoi(x,y,w,h); will give parameters for enclosing rectangle of oval Roi. ? crossed wire somewhere? :-) Come back with more explicit question perhaps. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-d-request@io.ece.drexel.edu Fri Nov 12 06:11 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id GAA18573; Fri, 12 Nov 1999 06:11:16 -0500 (EST) Date: Fri, 12 Nov 1999 06:11:16 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911121111.GAA18573@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #259 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/259 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 5098 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 259 Today's Topics: fourier analysis on a movie [ Dorothy Schafer To: nih-image@io.ece.drexel.edu Cc: wayne@codon.nih.gov Subject: fourier analysis on a movie Message-Id: Content-Type: text/plain; charset="us-ascii" ; format="flowed" Hi, We are interested in determining a measure of the fluxuation in fluorescence intensity over time, and in comparing that flux measurement on a cell to cell basis, so that data from different movies can be compared. A fourier transform analysis of the fluorescence intensity over time would do this. Within NIHImage, the standard FFT macros perform a two-dimensional fourier transform on a region of a single square image. Is there a macro that performs a three-dimensional fourier transform on a stack corresponding to a single region over time (as in a movie)? Thanks, Dorothy Schafer Dorothy Schafer Dept. of Cell Biology-Box 8228 Washington University School of Medicine 660 S. Euclid Ave. St. Louis, MO 63110 Ph: 314-362-0098 FAX: 314-362-0098 World Wide Web site: http://www.cooperlab.wustl.edu/ for FedEx/UPS Shipping: Dept. of Cell Biology Washington University School of Medicine 4566 Scott Ave. St. Louis, MO 63110 ------------------------------ Date: Thu, 11 Nov 1999 16:10:54 -0400 From: Seth Budick To: nih-image@io.ece.drexel.edu Subject: Re: crop Message-Id: Content-Type: text/plain; charset="us-ascii" >At 08:31 PM 11/9/99 -0600, you wrote: >>Hi, >>I have acquired a movie of TIFF stored in stack. >>However, I only need one fixed portion of the image to analyze for the >>area image by image. Do you have any suggestions of how to do this? Does >>NIH-Image has a crop tool which can apply to the stack? Or I need to write >>a special Macro to crop the rectangular are I need for analysis? >>Also, I wonder if you have any method for meausring area of dendrites? > >The "Crop and Scale-fast" macro in the "Stacks" macro file will crop a >stack. Or you can use the Image/Crop command in ImageJ. > >-wayne Is there a way to do this for oval ROIs since Scale and Rotate requires rectangular ROIs? Thanks, Seth ________________________________ Seth Budick Biology Department 414 Mugar Hall Northeastern University Boston, MA 02115 Phone: (617) 373-4034 e-mail: sbudick1@swarthmore.edu http://www.omalleylab.neu.edu ------------------------------ Date: Fri, 12 Nov 1999 9:55:26 +1100 From: GJOSS@rna.bio.mq.edu.au To: sbudick1@swarthmore.edu, nih-image@io.ece.drexel.edu Subject: Re: crop Message-ID: <128AB0A4488@rna.bio.mq.edu.au> >Date: Thu, 11 Nov 1999 16:10:54 -0400 >To: nih-image@io.ece.drexel.edu >From: Seth Budick >Subject: Re: crop > >>At 08:31 PM 11/9/99 -0600, you wrote: >>>Hi, >>>I have acquired a movie of TIFF stored in stack. >>>However, I only need one fixed portion of the image to analyze for >the >>>area image by image. Do you have any suggestions of how to do this? >Does >>>NIH-Image has a crop tool which can apply to the stack? Or I need to >write >>>a special Macro to crop the rectangular are I need for analysis? >>>Also, I wonder if you have any method for meausring area of >dendrites? >> >>The "Crop and Scale-fast" macro in the "Stacks" macro file will crop a >>stack. Or you can use the Image/Crop command in ImageJ. >> >>-wayne > >Is there a way to do this for oval ROIs since Scale and Rotate requires >rectangular ROIs? >Thanks, >Seth > >________________________________ >Seth Budick >Biology Department >414 Mugar Hall >Northeastern University >Boston, MA 02115 > >Phone: (617) 373-4034 >e-mail: sbudick1@swarthmore.edu >http://www.omalleylab.neu.edu > 'Croping a movie' normally implies that you wish to produce a smaller frame movie as the output. All movies, stacks, windows, frames are rectangular. You can make oval ROIs within the frame of a window or stack. If made on a stack, it is retained on the stack as you turn to each frame to process through the stack (by selectSlice?) unless the process is one that removes the ROI. In this case you need to reinstate the ROI by any of a number of techniques including use of "restoreRoi","save as(outline" etc. getRoi(x,y,w,h); will give parameters for enclosing rectangle of oval Roi. ? crossed wire somewhere? :-) Come back with more explicit question perhaps. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia -------------------------------- End of nih-image-d Digest V99 Issue #259 **************************************** From nih-image-request@io.ece.drexel.edu Fri Nov 12 16:14 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA15269; Fri, 12 Nov 1999 16:14:08 -0500 (EST) Resent-Date: Fri, 12 Nov 1999 16:14:08 -0500 (EST) Message-Id: <382C7F20.8F48FB62@maroon.tc.umn.edu> Date: Fri, 12 Nov 1999 14:58:01 -0600 From: "Michael J. Herron" Reply-To: Michael J Herron Organization: U of MN, Deaprtment of Medicine,Infectious Diseases X-Mailer: Mozilla 4.7 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu, GJOSS@rna.bio.mq.edu.au CC: sbudick1@swarthmore.edu Subject: NIH Image list References: <128AB0A4488@rna.bio.mq.edu.au> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"7z9ck1.0.Q_2.Mz7Bu"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1938 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 2563 Status: O Is there a problem with the list? I have seen VERY few posts for the last week. I asked a question abour SPOT cameras and got no reply... Just curious. Mike GJOSS@rna.bio.mq.edu.au wrote: > > >Date: Thu, 11 Nov 1999 16:10:54 -0400 > >To: nih-image@io.ece.drexel.edu > >From: Seth Budick > >Subject: Re: crop > > > >>At 08:31 PM 11/9/99 -0600, you wrote: > >>>Hi, > >>>I have acquired a movie of TIFF stored in stack. > >>>However, I only need one fixed portion of the image to analyze for > >the > >>>area image by image. Do you have any suggestions of how to do this? > >Does > >>>NIH-Image has a crop tool which can apply to the stack? Or I need to > >write > >>>a special Macro to crop the rectangular are I need for analysis? > >>>Also, I wonder if you have any method for meausring area of > >dendrites? > >> > >>The "Crop and Scale-fast" macro in the "Stacks" macro file will crop a > >>stack. Or you can use the Image/Crop command in ImageJ. > >> > >>-wayne > > > >Is there a way to do this for oval ROIs since Scale and Rotate requires > >rectangular ROIs? > >Thanks, > >Seth > > > >________________________________ > >Seth Budick > >Biology Department > >414 Mugar Hall > >Northeastern University > >Boston, MA 02115 > > > >Phone: (617) 373-4034 > >e-mail: sbudick1@swarthmore.edu > >http://www.omalleylab.neu.edu > > > > 'Croping a movie' normally implies that you wish to produce a smaller > frame movie as the output. All movies, stacks, windows, frames are > rectangular. > You can make oval ROIs within the frame of a window or stack. > If made on a stack, it is retained on the stack as you turn to each > frame to process through the stack (by selectSlice?) unless the process > is one that removes the ROI. > In this case you need to reinstate the ROI by any of a number of > techniques including use of "restoreRoi","save as(outline" etc. > > getRoi(x,y,w,h); will give parameters for enclosing rectangle of oval > Roi. > > ? crossed wire somewhere? :-) Come back with more explicit question > perhaps. > Greg Joss, > School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 > Macquarie University, Email gjoss@rna.bio.mq.edu.au > North Ryde, (Sydney,) NSW 2109, Australia -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ From nih-image-request@io.ece.drexel.edu Fri Nov 12 16:42 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA20878; Fri, 12 Nov 1999 16:42:24 -0500 (EST) Resent-Date: Fri, 12 Nov 1999 16:42:24 -0500 (EST) X-Sender: a_team@pop.dds.nl Message-Id: In-Reply-To: <199911121101.GAA16686@io.ece.drexel.edu> Mime-Version: 1.0 Date: Fri, 12 Nov 1999 20:52:28 +0100 To: nih-image@io.ece.drexel.edu From: "Anneke M.Th. Harbers and Ard Jonker" Subject: Re: crop Resent-Message-ID: <"yaUN4.0.RM4.FP8Bu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1939 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 408 Status: O >Is there a way to do this for oval ROIs since Scale and Rotate requires >rectangular ROIs? If you want to operate on a ROI only, save the outline and do in a loop an activate input stack, 'load outline', restore roi, copy, delete current slice, (switch from the input stack to the output stack now), add slice, select all, clear; paste and you will have put only an oval roi into the output image. ard From nih-image-request@io.ece.drexel.edu Fri Nov 12 16:54 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id QAA23063; Fri, 12 Nov 1999 16:54:34 -0500 (EST) Resent-Date: Fri, 12 Nov 1999 16:54:34 -0500 (EST) Date: Fri, 12 Nov 1999 16:38:11 -0500 From: Bill Christens-Barry Subject: Re: fourier analysis on a movie In-reply-to: <199911121111.GAA18593@io.ece.drexel.edu> To: nih-image@io.ece.drexel.edu, dorothy@cellbio.wustl.edu Message-id: MIME-version: 1.0 Content-transfer-encoding: 7BIT References: <199911121111.GAA18593@io.ece.drexel.edu> Resent-Message-ID: <"SB7653.0.vn4.DY8Bu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1940 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii; format=flowed Content-Length: 4028 Status: O >Date: Thu, 11 Nov 1999 10:48:26 -0600 >From: Dorothy Schafer >Subject: fourier analysis on a movie >To: nih-image@io.ece.drexel.edu >Cc: wayne@codon.nih.gov >Message-id: >Content-type: text/plain; format=flowed; charset=us-ascii >Content-transfer-encoding: 7BIT > >Hi, > > We are interested in determining a measure of the fluxuation >in fluorescence intensity over time, and in comparing that flux >measurement on a cell to cell basis, so that data from different >movies can be compared. A fourier transform analysis of the >fluorescence intensity over time would do this. Within NIHImage, >the standard FFT macros perform a two-dimensional fourier transform >on a region of a single square image. Is there a macro that >performs a three-dimensional fourier transform on a stack >corresponding to a single region over time (as in a movie)? > >Thanks, >Dorothy Schafer > >Dorothy Schafer >Dept. of Cell Biology-Box 8228 >Washington University School of Medicine >660 S. Euclid Ave. >St. Louis, MO 63110 > >Ph: 314-362-0098 >FAX: 314-362-0098 >World Wide Web site: http://www.cooperlab.wustl.edu/ > >for FedEx/UPS Shipping: >Dept. of Cell Biology >Washington University School of Medicine >4566 Scott Ave. >St. Louis, MO 63110 > Dorothy, This is something that I've done before, but not in NIH Image, and I don't know of any current implementation using Image. It would be straightforward to do, with qualifications. These are: i) NIH Image is set up to do only 2-d FFTs, not 1-d FFTs (it's funny to think of this extra dimensionality as an "only", as it's common that time domain packages can only do 1-d FFTs). So, whatever Image you use to represent your time domain data you will have to encode it into a 2-d image. ii) The image representing the time domain data will have to have 2^N x 2^N pixels, N an integer, as does any FFT-able image in NIH Image. In light of this, you can imagine building an image in which successive columns represent temporally successive measurements. So, for example, you could come up with a single measurement (the integrated intensity of a cell, say) from the first image in your time series stack and fill each pixel in the first column of your time domain image with this value. Fill the second column of your time domain image with values obtained from the second image in your time series stack. Do this until all 2^N columns are full. Now you can do your FFT operations. The power spectrum will appear as a function of column number in the result (i.e. it will vary horizontally in the transformed image). This process would have to be repeated for each cell or object of interest. Something to think about is the normalization of the signal used to represent different cells. If you are only interested in comparing the time dependence of each cell's fluorescence, then normalization is no problem. If you want to compare the intensities in spectral bands among cells, then you'll have to use a normalization. A consequence of this method is the third qualification: iii) Your time series stack will have to have at least 2^N usable, consecutive frames to fill the 2^N columns. BTW, this approach ignores/suppresses all y-dependence and thus the FFT will show only a DC component in y. With a little more sophistication, you might be able to exploit the y-dimension to look at some other dependence (e.g. wavelength). If you do this, each row will indicate the power spectrum of the corresponding wavelength. In a study of cell motility, our group encoded cell boundary shapes as 2-d chain curves, and then took temporal FFTs of the spatial FFTs in order to characterize the various types of motile behavior that contributed to the cell's overall motion. Something like this will require that the stack be some 2^N x 2^N x 2^M in size. Hope this helps, and let me know if you'd like more on this. Bill Christens-Barry wacb@aplcomm.jhuapl.edu 443-778-6253 From nih-image-request@io.ece.drexel.edu Sat Nov 13 05:16 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id FAA02962; Sat, 13 Nov 1999 05:16:26 -0500 (EST) Resent-Date: Sat, 13 Nov 1999 05:16:26 -0500 (EST) Date: Sat, 13 Nov 1999 05:05:00 -0500 (EST) From: nih-image-owner@io.ece.drexel.edu Message-Id: <199911131005.FAA00583@io.ece.drexel.edu> To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Resent-Message-ID: <"3AACv1.0.U9.EVJBu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1941 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 3783 Status: O NIH-image Mailing List Help --------------------------- ********************** * UNSUBSCRIBE HERE * ********************** nih-image-d-request@biomed.drexel.edu - Digest Version nih-image-request@biomed.drexel.edu - Non-digest Version A digest means that you get all of the list email for any day in one email. The alternative is when you get one email for every post to the list. *Note: sending "unsubscribe" to the everybody on the list makes you look really silly. Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- From nih-image-d-request@io.ece.drexel.edu Sat Nov 13 19:50 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA05207; Sat, 13 Nov 1999 19:50:34 -0500 (EST) Date: Sat, 13 Nov 1999 19:50:34 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911140050.TAA05207@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #260 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/260 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 13557 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 260 Today's Topics: NIH Image list [ "Michael J. Herron" To: nih-image@io.ece.drexel.edu, GJOSS@rna.bio.mq.edu.au CC: sbudick1@swarthmore.edu Subject: NIH Image list Message-Id: <382C7F20.8F48FB62@maroon.tc.umn.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Is there a problem with the list? I have seen VERY few posts for the last week. I asked a question abour SPOT cameras and got no reply... Just curious. Mike GJOSS@rna.bio.mq.edu.au wrote: > > >Date: Thu, 11 Nov 1999 16:10:54 -0400 > >To: nih-image@io.ece.drexel.edu > >From: Seth Budick > >Subject: Re: crop > > > >>At 08:31 PM 11/9/99 -0600, you wrote: > >>>Hi, > >>>I have acquired a movie of TIFF stored in stack. > >>>However, I only need one fixed portion of the image to analyze for > >the > >>>area image by image. Do you have any suggestions of how to do this? > >Does > >>>NIH-Image has a crop tool which can apply to the stack? Or I need to > >write > >>>a special Macro to crop the rectangular are I need for analysis? > >>>Also, I wonder if you have any method for meausring area of > >dendrites? > >> > >>The "Crop and Scale-fast" macro in the "Stacks" macro file will crop a > >>stack. Or you can use the Image/Crop command in ImageJ. > >> > >>-wayne > > > >Is there a way to do this for oval ROIs since Scale and Rotate requires > >rectangular ROIs? > >Thanks, > >Seth > > > >________________________________ > >Seth Budick > >Biology Department > >414 Mugar Hall > >Northeastern University > >Boston, MA 02115 > > > >Phone: (617) 373-4034 > >e-mail: sbudick1@swarthmore.edu > >http://www.omalleylab.neu.edu > > > > 'Croping a movie' normally implies that you wish to produce a smaller > frame movie as the output. All movies, stacks, windows, frames are > rectangular. > You can make oval ROIs within the frame of a window or stack. > If made on a stack, it is retained on the stack as you turn to each > frame to process through the stack (by selectSlice?) unless the process > is one that removes the ROI. > In this case you need to reinstate the ROI by any of a number of > techniques including use of "restoreRoi","save as(outline" etc. > > getRoi(x,y,w,h); will give parameters for enclosing rectangle of oval > Roi. > > ? crossed wire somewhere? :-) Come back with more explicit question > perhaps. > Greg Joss, > School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 > Macquarie University, Email gjoss@rna.bio.mq.edu.au > North Ryde, (Sydney,) NSW 2109, Australia -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ ------------------------------ Date: Fri, 12 Nov 1999 20:52:28 +0100 From: "Anneke M.Th. Harbers and Ard Jonker" To: nih-image@io.ece.drexel.edu Subject: Re: crop Message-Id: Content-Type: text/plain; charset="us-ascii" >Is there a way to do this for oval ROIs since Scale and Rotate requires >rectangular ROIs? If you want to operate on a ROI only, save the outline and do in a loop an activate input stack, 'load outline', restore roi, copy, delete current slice, (switch from the input stack to the output stack now), add slice, select all, clear; paste and you will have put only an oval roi into the output image. ard ------------------------------ Date: Fri, 12 Nov 1999 16:38:11 -0500 From: Bill Christens-Barry To: nih-image@io.ece.drexel.edu, dorothy@cellbio.wustl.edu Subject: Re: fourier analysis on a movie Message-id: Content-type: text/plain; charset=us-ascii; format=flowed Content-transfer-encoding: 7BIT >Date: Thu, 11 Nov 1999 10:48:26 -0600 >From: Dorothy Schafer >Subject: fourier analysis on a movie >To: nih-image@io.ece.drexel.edu >Cc: wayne@codon.nih.gov >Message-id: >Content-type: text/plain; format=flowed; charset=us-ascii >Content-transfer-encoding: 7BIT > >Hi, > > We are interested in determining a measure of the fluxuation >in fluorescence intensity over time, and in comparing that flux >measurement on a cell to cell basis, so that data from different >movies can be compared. A fourier transform analysis of the >fluorescence intensity over time would do this. Within NIHImage, >the standard FFT macros perform a two-dimensional fourier transform >on a region of a single square image. Is there a macro that >performs a three-dimensional fourier transform on a stack >corresponding to a single region over time (as in a movie)? > >Thanks, >Dorothy Schafer > >Dorothy Schafer >Dept. of Cell Biology-Box 8228 >Washington University School of Medicine >660 S. Euclid Ave. >St. Louis, MO 63110 > >Ph: 314-362-0098 >FAX: 314-362-0098 >World Wide Web site: http://www.cooperlab.wustl.edu/ > >for FedEx/UPS Shipping: >Dept. of Cell Biology >Washington University School of Medicine >4566 Scott Ave. >St. Louis, MO 63110 > Dorothy, This is something that I've done before, but not in NIH Image, and I don't know of any current implementation using Image. It would be straightforward to do, with qualifications. These are: i) NIH Image is set up to do only 2-d FFTs, not 1-d FFTs (it's funny to think of this extra dimensionality as an "only", as it's common that time domain packages can only do 1-d FFTs). So, whatever Image you use to represent your time domain data you will have to encode it into a 2-d image. ii) The image representing the time domain data will have to have 2^N x 2^N pixels, N an integer, as does any FFT-able image in NIH Image. In light of this, you can imagine building an image in which successive columns represent temporally successive measurements. So, for example, you could come up with a single measurement (the integrated intensity of a cell, say) from the first image in your time series stack and fill each pixel in the first column of your time domain image with this value. Fill the second column of your time domain image with values obtained from the second image in your time series stack. Do this until all 2^N columns are full. Now you can do your FFT operations. The power spectrum will appear as a function of column number in the result (i.e. it will vary horizontally in the transformed image). This process would have to be repeated for each cell or object of interest. Something to think about is the normalization of the signal used to represent different cells. If you are only interested in comparing the time dependence of each cell's fluorescence, then normalization is no problem. If you want to compare the intensities in spectral bands among cells, then you'll have to use a normalization. A consequence of this method is the third qualification: iii) Your time series stack will have to have at least 2^N usable, consecutive frames to fill the 2^N columns. BTW, this approach ignores/suppresses all y-dependence and thus the FFT will show only a DC component in y. With a little more sophistication, you might be able to exploit the y-dimension to look at some other dependence (e.g. wavelength). If you do this, each row will indicate the power spectrum of the corresponding wavelength. In a study of cell motility, our group encoded cell boundary shapes as 2-d chain curves, and then took temporal FFTs of the spatial FFTs in order to characterize the various types of motile behavior that contributed to the cell's overall motion. Something like this will require that the stack be some 2^N x 2^N x 2^M in size. Hope this helps, and let me know if you'd like more on this. Bill Christens-Barry wacb@aplcomm.jhuapl.edu 443-778-6253 ------------------------------ Date: Sat, 13 Nov 1999 05:05:00 -0500 (EST) From: nih-image-owner@io.ece.drexel.edu To: nih-image@io.ece.drexel.edu Subject: ADMIN: list commands Message-Id: <199911131005.FAA00583@io.ece.drexel.edu> NIH-image Mailing List Help --------------------------- ********************** * UNSUBSCRIBE HERE * ********************** nih-image-d-request@biomed.drexel.edu - Digest Version nih-image-request@biomed.drexel.edu - Non-digest Version A digest means that you get all of the list email for any day in one email. The alternative is when you get one email for every post to the list. *Note: sending "unsubscribe" to the everybody on the list makes you look really silly. Subscribe --------- To subscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: subscribe Subscribe to Digest ------------------- To subscribe to a digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "Subscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: subscribe Unsubscribe ----------- To unsubscribe to the list, send E-mail to nih-image-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-request@biomed.drexel.edu Subject: unsubscribe Unsubscribe to Digest -------------------- To unsubscribe to digested version of the list, send E-mail to nih-image-d-request@biomed.drexel.edu. The Subject of the message should contain "unsubscribe". As in: To: nih-image-d-request@biomed.drexel.edu Subject: unsubscribe Archive ------- Every submission sent to this list is archived. Following are the commands used to access the archive. Send Email to nih-image-request@biomed.drexel.edu with the command in the Subject line or in the body of the message. Tips by Henry M. Thomas, 0) Remember that Archive is case-sensitive. 1) Use the proper address for the Archive nih-image-request@biomed.drexel.edu 2) title the Subject line of your email archive 3) turn off (or don't send) your email Signature if you have one 4) In the body of the email write ls latest 5) Send it. latest is a directory containing the latest 50 files. The ls command returns you a list of the filenumbers of the current 50 files. (currently in the 800's). so latest/862 is a filename. 6) Send a second email get latest/862 or whatever filenumber you need 7) But you probaly want a batch. If you want more than 16, start the body of the email with archive maxfiles 35 or however many you want then use Unix metacharacters to get more than one file. * matches any string. ? matches any single character. []'s matches any single character shown in the brackets. get latest/* all 50 get latest/8[3-6]? all from 830 to 869 8) To search for text (e.g. the word color) in all the files in the directory latest use egrep egrep color latest/* then use get to get the files you want. This archive server knows the following commands: get filename ... ls directory ... egrep case_insensitive_regular_expression filename ... maxfiles nnn version quit Aliases for 'get': send, sendme, getme, gimme, retrieve, mail Aliases for 'ls': dir, directory, list, show Aliases for 'egrep': search, grep, fgrep, find Aliases for 'quit': exit Lines starting with a '#' are ignored. Multiple commands per mail are allowed. Setting maxfiles to zero will remove the limit (to protect you against yourself no more than maxfiles files will be returned per request). Egrep supports most common flags. If you append a non-standard signature, you should use the quit command to prevent the archive server from interpreting the signature. Examples: ls latest get latest/12 egrep some.word latest/* -- ------------------------------ Date: Sat, 13 Nov 1999 18:38:20 -0600 (CST) From: Ken Johnson To: NIH Image User Group Subject: Image Adjustment-contrast Message-ID: Content-Type: TEXT/PLAIN; charset=US-ASCII Imagers I have been using ImageJ to create stacks of images for exports into movie format with a reasonable success. However, in some cases I wish to adjust the contrast/brightness on these stacks, but wish to apply the same enhancements to each image within the stack. When I use the Image adjust feature I have to apply this to each image indiviually, and ImageJ does not seem to like me doing this. Is there an easy way to apply a single criteria to the entire stack?? Thanks for your help! Kenneth G. Johnson, Ph.D. Washington University Medical School St. Louis, MO. -------------------------------- End of nih-image-d Digest V99 Issue #260 **************************************** From nih-image-request@io.ece.drexel.edu Sat Nov 13 19:52 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id TAA05531; Sat, 13 Nov 1999 19:52:01 -0500 (EST) Resent-Date: Sat, 13 Nov 1999 19:52:01 -0500 (EST) Date: Sat, 13 Nov 1999 18:38:20 -0600 (CST) From: Ken Johnson X-Sender: kenjohn@pathbox.wustl.edu To: NIH Image User Group Subject: Image Adjustment-contrast Message-ID: MIME-Version: 1.0 Resent-Message-ID: <"_lP45.0.9c.4IWBu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1942 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Length: 577 Status: O Imagers I have been using ImageJ to create stacks of images for exports into movie format with a reasonable success. However, in some cases I wish to adjust the contrast/brightness on these stacks, but wish to apply the same enhancements to each image within the stack. When I use the Image adjust feature I have to apply this to each image indiviually, and ImageJ does not seem to like me doing this. Is there an easy way to apply a single criteria to the entire stack?? Thanks for your help! Kenneth G. Johnson, Ph.D. Washington University Medical School St. Louis, MO. From nih-image-request@io.ece.drexel.edu Sun Nov 14 18:37 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id SAA22687; Sun, 14 Nov 1999 18:37:26 -0500 (EST) Resent-Date: Sun, 14 Nov 1999 18:37:26 -0500 (EST) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: herro001@maroon.tc.umn.edu, nih-image@io.ece.drexel.edu Date: Mon, 15 Nov 1999 10:31:29 +1100 Subject: Re: NIH Image list / archive access problems Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <1714AF44901@rna.bio.mq.edu.au> Resent-Message-ID: <"uWuQ51.0.N15.CKqBu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1943 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 2078 Status: O >Date: Fri, 12 Nov 1999 14:58:01 -0600 >From: "Michael J. Herron" >To: nih-image@io.ece.drexel.edu, GJOSS@rna.bio.mq.edu.au >CC: sbudick1@swarthmore.edu >Subject: NIH Image list > >Is there a problem with the list? I have seen VERY few posts for the >last week. >I asked a question abour SPOT cameras and got no reply... > >Just curious. > >Mike Mike, There haven't been any apparent problems with the "normal" operation of the list that I have noticed. I did receive your post "Spot aquisition". (I had nothing useful to contribute). On the other hand, while not wanting to "look a gift horse in the mouth": The NIH-Image maillist quality has suffered considerably over the past year as a result of the move to the new administration. Access to the archives via email exchange is quite unwieldy by todays standards, particularly when compared with the ready web access and search capabilities offered by ImageJ at http://list.nih.gov/archives/imagej.html Confocal at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal I have previously tried to discuss this failing with administrators but raised little interest. I can understand Wayne wishing to transfer his efforts to ImageJ developments but ImageJ is not yet ready to address the same needs as NIH-Image has done so successfully. I cannot understand why NIH doesn't give the NIH-Image mail list the same support it used to. Access to the complete archive has evaporated. The lack of easy and comprehensive access to archives results in a much reduced value of the maillist to researchers, many faqs (frequently asked questions) being repeated, dropping the value and quality of the maillist for all concerned in comparision to previous years. This is despite a much increased population of potential correspondents and interested researchers. Any suggestions for improvement? Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Mon Nov 15 02:58 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id CAA11635; Mon, 15 Nov 1999 02:58:46 -0500 (EST) Resent-Date: Mon, 15 Nov 1999 02:58:46 -0500 (EST) From: GJOSS@rna.bio.mq.edu.au Organization: Dept. of Biological Sciences To: dorothy@cellbio.wustl.edu, wacb@aplcomm.jhuapl.edu, nih-image@io.ece.drexel.edu Date: Mon, 15 Nov 1999 18:45:52 +1100 Subject: Re: fourier analysis on a movie Priority: normal X-mailer: Pegasus Mail/Mac (v2.2.1) Message-ID: <17988B37744@rna.bio.mq.edu.au> Resent-Message-ID: <"uJq_03.0.oB2.vcxBu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1944 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text Content-Length: 2012 Status: O >Date: Fri, 12 Nov 1999 16:38:11 -0500 >From: Bill Christens-Barry >Subject: Re: fourier analysis on a movie >To: nih-image@io.ece.drexel.edu, dorothy@cellbio.wustl.edu > >>Date: Thu, 11 Nov 1999 10:48:26 -0600 >>From: Dorothy Schafer >>Subject: fourier analysis on a movie >>To: nih-image@io.ece.drexel.edu >> >> We are interested in determining a measure of the fluxuation >>in fluorescence intensity over time, and in comparing that flux >>measurement on a cell to cell basis, so that data from different >>movies can be compared. A fourier transform analysis of the >>fluorescence intensity over time would do this.................. > >Dorothy, > >This is something that I've done before,................, and let me know if you'd like more on this. > >Bill Christens-Barry I am interested in Dorothy's question and Bill's response so please dont take discussion off the list. Dorothy, What is the timescale of your fluorecence movie? ie what is the integration time per frame ? what is frame rate? are these confocal slices or whole cell projections? The 2^n dimensional restrictions imposed by Fourier transforms and refered to by Bill would seem to me to be problematic. Even with a 2^n^3 dataset, the unclear relationship between the spatial and time scales adds further confusion to how I could interpret the result of a '3D fourier transform'. Bill, As I interpreted your suggestion, you would be simply batching a set of 2^n time series of integrated cell intensites into a 2D FFT for processing; using the results the same as if you had done 2^n independant 1D FT's. I suspect the chain code motility case did successive 2D FFT rather than anything 3D. Would you explain how you would get anything meaningful out of the process please. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia From nih-image-request@io.ece.drexel.edu Mon Nov 15 07:54 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id HAA29252; Mon, 15 Nov 1999 07:54:30 -0500 (EST) Resent-Date: Mon, 15 Nov 1999 07:54:30 -0500 (EST) Message-Id: <382FFF19.9104ECAF@maroon.tc.umn.edu> Date: Mon, 15 Nov 1999 06:44:44 -0600 From: "Michael J. Herron" Reply-To: Michael J Herron Organization: U of MN, Deaprtment of Medicine,Infectious Diseases X-Mailer: Mozilla 4.7 (Macintosh; I; PPC) X-Accept-Language: en MIME-Version: 1.0 To: nih-image@io.ece.drexel.edu Subject: Re: NIH Image list / archive access problems References: <1714AF44901@rna.bio.mq.edu.au> Content-Transfer-Encoding: 7bit Resent-Message-ID: <"U3F7W2.0.8X6.8z_Bu"@io> Resent-From: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1945 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset=us-ascii Content-Length: 2247 Status: O I agree. I have tried to get into the current archive twice, before giving up both times. If it is too difficult to build a little (friendly) search engine, perhaps the list administrator could make the entire text file available? If it was available fo download in a couple of zipped or stuffed chunks, anyone could search it from a word processor. Mike GJOSS@rna.bio.mq.edu.au wrote: > . . . > while not wanting to "look a gift horse in the mouth": > > The NIH-Image maillist quality has suffered considerably over the past > year as a result of the move to the new administration. Access to the > archives via email exchange is quite unwieldy by todays standards, > particularly when compared with the ready web access and search > capabilities offered by > ImageJ at http://list.nih.gov/archives/imagej.html > Confocal at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > I have previously tried to discuss this failing with administrators but > raised little interest. I can understand Wayne wishing to transfer his > efforts to ImageJ developments but ImageJ is not yet ready to address > the same needs as NIH-Image has done so successfully. I cannot > understand why NIH doesn't give the NIH-Image mail list the same support > it used to. Access to the complete archive has evaporated. > > The lack of easy and comprehensive access to archives results in a much > reduced value of the maillist to researchers, many faqs (frequently > asked questions) being repeated, dropping the value and quality of the > maillist for all concerned in comparision to previous years. > This is despite a much increased population of potential correspondents > and interested researchers. > > Any suggestions for improvement? > > Greg Joss, > School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 > Macquarie University, Email gjoss@rna.bio.mq.edu.au > North Ryde, (Sydney,) NSW 2109, Australia -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ From nih-image-d-request@io.ece.drexel.edu Mon Nov 15 10:57 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id KAA29314; Mon, 15 Nov 1999 10:57:16 -0500 (EST) Date: Mon, 15 Nov 1999 10:57:16 -0500 (EST) From: nih-image-d-request@io.ece.drexel.edu Message-Id: <199911151557.KAA29314@io.ece.drexel.edu> Subject: nih-image-d Digest V99 #261 X-Loop: nih-image-d@biomed.drexel.edu X-Mailing-List: archive/volume99/261 Precedence: list MIME-Version: 1.0 To: nih-image-d@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu Content-Type: multipart/digest; boundary="----------------------------" Content-Length: 8941 Status: O ------------------------------ Content-Type: text/plain nih-image-d Digest Volume 99 : Issue 261 Today's Topics: Re: NIH Image list / archive access [ GJOSS@rna.bio.mq.edu.au ] Re: fourier analysis on a movie [ GJOSS@rna.bio.mq.edu.au ] Re: NIH Image list / archive access [ "Michael J. Herron" ] ------------------------------ Date: Mon, 15 Nov 1999 10:31:29 +1100 From: GJOSS@rna.bio.mq.edu.au To: herro001@maroon.tc.umn.edu, nih-image@io.ece.drexel.edu Subject: Re: NIH Image list / archive access problems Message-ID: <1714AF44901@rna.bio.mq.edu.au> >Date: Fri, 12 Nov 1999 14:58:01 -0600 >From: "Michael J. Herron" >To: nih-image@io.ece.drexel.edu, GJOSS@rna.bio.mq.edu.au >CC: sbudick1@swarthmore.edu >Subject: NIH Image list > >Is there a problem with the list? I have seen VERY few posts for the >last week. >I asked a question abour SPOT cameras and got no reply... > >Just curious. > >Mike Mike, There haven't been any apparent problems with the "normal" operation of the list that I have noticed. I did receive your post "Spot aquisition". (I had nothing useful to contribute). On the other hand, while not wanting to "look a gift horse in the mouth": The NIH-Image maillist quality has suffered considerably over the past year as a result of the move to the new administration. Access to the archives via email exchange is quite unwieldy by todays standards, particularly when compared with the ready web access and search capabilities offered by ImageJ at http://list.nih.gov/archives/imagej.html Confocal at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal I have previously tried to discuss this failing with administrators but raised little interest. I can understand Wayne wishing to transfer his efforts to ImageJ developments but ImageJ is not yet ready to address the same needs as NIH-Image has done so successfully. I cannot understand why NIH doesn't give the NIH-Image mail list the same support it used to. Access to the complete archive has evaporated. The lack of easy and comprehensive access to archives results in a much reduced value of the maillist to researchers, many faqs (frequently asked questions) being repeated, dropping the value and quality of the maillist for all concerned in comparision to previous years. This is despite a much increased population of potential correspondents and interested researchers. Any suggestions for improvement? Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Mon, 15 Nov 1999 18:45:52 +1100 From: GJOSS@rna.bio.mq.edu.au To: dorothy@cellbio.wustl.edu, wacb@aplcomm.jhuapl.edu, nih-image@io.ece.drexel.edu Subject: Re: fourier analysis on a movie Message-ID: <17988B37744@rna.bio.mq.edu.au> >Date: Fri, 12 Nov 1999 16:38:11 -0500 >From: Bill Christens-Barry >Subject: Re: fourier analysis on a movie >To: nih-image@io.ece.drexel.edu, dorothy@cellbio.wustl.edu > >>Date: Thu, 11 Nov 1999 10:48:26 -0600 >>From: Dorothy Schafer >>Subject: fourier analysis on a movie >>To: nih-image@io.ece.drexel.edu >> >> We are interested in determining a measure of the fluxuation >>in fluorescence intensity over time, and in comparing that flux >>measurement on a cell to cell basis, so that data from different >>movies can be compared. A fourier transform analysis of the >>fluorescence intensity over time would do this.................. > >Dorothy, > >This is something that I've done before,................, and let me know if you'd like more on this. > >Bill Christens-Barry I am interested in Dorothy's question and Bill's response so please dont take discussion off the list. Dorothy, What is the timescale of your fluorecence movie? ie what is the integration time per frame ? what is frame rate? are these confocal slices or whole cell projections? The 2^n dimensional restrictions imposed by Fourier transforms and refered to by Bill would seem to me to be problematic. Even with a 2^n^3 dataset, the unclear relationship between the spatial and time scales adds further confusion to how I could interpret the result of a '3D fourier transform'. Bill, As I interpreted your suggestion, you would be simply batching a set of 2^n time series of integrated cell intensites into a 2D FFT for processing; using the results the same as if you had done 2^n independant 1D FT's. I suspect the chain code motility case did successive 2D FFT rather than anything 3D. Would you explain how you would get anything meaningful out of the process please. Greg Joss, School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 Macquarie University, Email gjoss@rna.bio.mq.edu.au North Ryde, (Sydney,) NSW 2109, Australia ------------------------------ Date: Mon, 15 Nov 1999 06:44:44 -0600 From: "Michael J. Herron" To: nih-image@io.ece.drexel.edu Subject: Re: NIH Image list / archive access problems Message-Id: <382FFF19.9104ECAF@maroon.tc.umn.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I agree. I have tried to get into the current archive twice, before giving up both times. If it is too difficult to build a little (friendly) search engine, perhaps the list administrator could make the entire text file available? If it was available fo download in a couple of zipped or stuffed chunks, anyone could search it from a word processor. Mike GJOSS@rna.bio.mq.edu.au wrote: > . . .> while not wanting to "look a gift horse in the mouth": > > The NIH-Image maillist quality has suffered considerably over the past > year as a result of the move to the new administration. Access to the > archives via email exchange is quite unwieldy by todays standards, > particularly when compared with the ready web access and search > capabilities offered by > ImageJ at http://list.nih.gov/archives/imagej.html > Confocal at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal > > I have previously tried to discuss this failing with administrators but > raised little interest. I can understand Wayne wishing to transfer his > efforts to ImageJ developments but ImageJ is not yet ready to address > the same needs as NIH-Image has done so successfully. I cannot > understand why NIH doesn't give the NIH-Image mail list the same support > it used to. Access to the complete archive has evaporated. > > The lack of easy and comprehensive access to archives results in a much > reduced value of the maillist to researchers, many faqs (frequently > asked questions) being repeated, dropping the value and quality of the > maillist for all concerned in comparision to previous years. > This is despite a much increased population of potential correspondents > and interested researchers. > > Any suggestions for improvement? > > Greg Joss, > School of Biological Sciences, Phone:(61)(2) 9850 8212 Fax: 9850 8174 > Macquarie University, Email gjoss@rna.bio.mq.edu.au > North Ryde, (Sydney,) NSW 2109, Australia -- _________________________________________________ / Michael J. Herron / / U of MN,Medicine/Infectious Diseases / / herro001@maroon.tc.umn.edu / / http://128.101.243.213 / /_____________________________________________/ ------------------------------ Date: Mon, 15 Nov 1999 11:53:51 -0400 From: Wayne Rasband To: nih-image@io.ece.drexel.edu Cc: imagej@list.nih.gov Subject: Re: Image Adjustment-contrast Message-Id: Content-Type: text/plain; charset="us-ascii" >Imagers > >I have been using ImageJ to create stacks of images for exports into movie >format with a reasonable success. However, in some cases I wish to adjust >the contrast/brightness on these stacks, but wish to apply the same >enhancements to each image within the stack. When I use the Image adjust >feature I have to apply this to each image indiviually, and ImageJ does >not seem to like me doing this. Is there an easy way to apply a single >criteria to the entire stack?? In ImageJ 1.10, clicking "Apply" in the Adjust Contrast window applies the LUT to the entire stack. In v1.11, it will give you the choice of applying the LUT to the entire stack or only to the current slice. Note that the "Apply" button only works with 8-bit stacks. To apply the LUT to an entire stack in NIH Image, use the "Apply LUT" macro in the Stacks macro file. -wayne -------------------------------- End of nih-image-d Digest V99 Issue #261 **************************************** From nih-image-request@io.ece.drexel.edu Mon Nov 15 10:59 EST 1999 Received: (from lists@localhost) by io.ece.drexel.edu (8.8.8/8.8.8) id KAA29706; Mon, 15 Nov 1999 10:59:14 -0500 (EST) Resent-Date: Mon, 15 Nov 1999 10:59:14 -0500 (EST) Message-Id: In-Reply-To: <3.0.5.32.19991114232835.00aa0cf0@codon.nih.gov> Mime-Version: 1.0 Date: Mon, 15 Nov 1999 11:53:51 -0400 To: nih-image@io.ece.drexel.edu From: Wayne Rasband Subject: Re: Image Adjustment-contrast Cc: imagej@list.nih.gov Resent-Message-ID: <"OzYVM.0.lW6.Re2Cu"@io> Resent-From: nih-image@io.ece.drexel.edu Reply-To: nih-image@io.ece.drexel.edu X-Mailing-List: archive/latest/1946 X-Loop: nih-image@biomed.drexel.edu Precedence: list Resent-Sender: nih-image-request@io.ece.drexel.edu Content-Type: text/plain; charset="us-ascii" Content-Length: 863 Status: O >Imagers > >I have been using ImageJ to create stacks of images for exports into movie >format with a reasonable success. However, in some cases I wish to adjust >the contrast/brightness on these stacks, but wish to apply the same >enhancements to each image within the stack. When I use the Image adjust >feature I have to apply this to each image indiviually, and ImageJ does >not seem to like me doing this. Is there an easy way to apply a single >criteria to the entire stack?? In ImageJ 1.10, clicking "Apply" in the Adjust Contrast window applies the LUT to the entire stack. In v1.11, it will give you the choice of applying the LUT to the entire stack or only to the current slice. Note that the "Apply" button only works with 8-bit stacks. To apply the LUT to an entire stack in NIH Image, use the "Apply LUT" macro in the Stacks macro file. -wayne