IT-LIST Digest 17Topics covered in this issue include:  1) RE: 	by Don Wilcox   2) IT	by "henry.valk"   3) 1.23 Problems	by Squiers   4) frame grabbers	by ben   5) Color Imaging	by "S. Brent Dove"   6) Re: Fourier Transform	by "S. Brent Dove"   7) Re: Color Imaging	by Hal Hawkins   8) Re: Color Imaging	by Alan Hale   9) how to add scaning to my app ?	by gil@commtouch.co.il (gil) 10) Re: Color Imaging	by Gary Krause  11) ITSDK Zip file	by "S. Brent Dove" ----------------------------------------------------------------------Date: Wed, 3 Apr 1996 15:41:14 -0700From: Don Wilcox To: 'ImageTool List Server' Subject: RE: Message-ID: <01BB2212.82E015A0@slip9.xroads.com>----------From:  PH ARTES[SMTP:P.H.Artes@bradford.ac.uk]Sent:  Monday, April 01, 1996 5:32 AMTo:  wilcox@xroads.comHas anybody experience in doing fourier-transforms and frequency-domain=20filtering on images using ImageTool? Is there a pre-made tool box or=20would the process involve programming?ImageTool currently does not provide built-in support for fourier =transforms.  I have it on the list, but am still trying to decide how to =do it best within the framework of IT.  If you have C code that does a =fourier transform, you could make it into a plug-in with very little =effort.  The documentation on the plug-in format is available in the =SDK, which I am told is currently corrupted.Thanks,paulDon------------------------------Date: Fri, 05 Apr 1996 16:52:17 +1000From: "henry.valk" To: it-list@sparky.uthscsa.eduSubject: ITMessage-ID: <3164C321.4524@netinfo.com.au>I have been playing around with Image Tools V1.23 to evaluate its potential as processing software for an automated laboratory application and have observed the following;1. There is only one available calibration process which is applied to the whole image, ie, horizontal and vertical planes. Generally the spacial relationship in these two planes will be different due to the lens used, the pixel spacing on the CCD and the typical 4:3 aspect ratio of most imaging systems.If an image is calibrated in the horizontal plane, for example, then any off axis measurements made will be in error. 2. In addition to item 1, there is no facility for mapping out any non linearities during the calibration procedure,ie, if measurements are made in various parts of the image there is no way of accounting for linearity variations at the edge of the image compared to the centre of the image.3. There does not appear to be an automatic scroll option for making measurements of objects which extend beyond the screen perimeter, typically for zoomed areas of the image.4. There does not appear to be an area based histogram feature where the area is selected by the user (Also very usefull for colour identification).5. There is no facility to determine the distance between the rising edge and falling edge of a typical line histogram, ie, it would be nice to be able to measure something by applying a line histogram and defining a threshold to the data such that the distance between two consecutive threshold values is given based on the spacial calibration applied.6. As per a previous message, colour images are not yet adequately supported. In particular, a colour image can be separated to its primary constituent colours but no histogram option can be applied to the sub images. 7. Similarly, there is no option to classify a particular colour and find all objects in a colour image of the same colour (with an applied tolerance).8. IT does no appear to support automatic loading and processing of images. I believe that the source code is available though from which an embedded application may be written.Wish list : 	A very handy DLL would be one that reads barcode labels for automatic object identification. Either automatically identifying the barcode label type and providing the barcode number or one DLL for each barcode type.	Extension of the maths functions to perform volume calculations given object characteristic information, ie, objects of type "x" are assumed to be cylindrical, or spherical, etc, find volume of object. Similarly, object subtraction, ie, for all objects of type "x", assumed to be cylindrical, find volume, subtract object of type "y", assumed to be cylindrical, calculate total volume (handy for calculating the volume of the contents of a vessel knowing the characteristics of that vessel).	This could also be extended to conditional math functions whereby the volume of a particular coloured layer in a known vessel is determined by applying a condition to the previous example of "colour type".Please let me know if any of my observations are incorrect or if I am trying to use IT for something other than intended. Having said all this I would like to add that IT is good tool for image processing and will likely evolve into an excellent tool in the near future.Henry Valk------------------------------Date: Mon, 08 Apr 1996 11:16:45 -0400From: Squiers To: it-list@sparky.uthscsa.eduSubject: 1.23 ProblemsMessage-ID: <1.5.4b13.32.19960408151645.00684e98@gateway.cybernet.com>This is the first time I've tried porting images created in IT to theMacintosh...so these errors may have already been discussed...1) PICT format is unreadable by GraphicConverter 2.3.1.  This is a highly   reliable program, so I'm guessing it's a problem with the Macintosh   export in IT 1.232) .TIF format resolution should probably default to something like   72x72 DPI (for typical screen resolution, although, Windows just   deal in pixels - one can always hope it's close; it's a Mac "standard").   One file I created via a Window's Bitmap was read into the Mac as   having a resolution of 26x72 DPI and had to be corrected.  My guess would   be that this would be a nice feature to be able to control manually.  If   there is a way to do this, I'd be grateful to learn of it.3) Please get the version numbers on stuff organized :)  Your web page says   1.2, and the readme files in the install says 1.1.  I'd like to see   a version number on the zip file (i.e., IT123.ZIP) to explicitely give   versions.  I realize this takes a little more time because the web page   has to be updated for the new filename everytime an update it completed,   but I think it'll be worth the effort (as least from my point of view ;) )Enough babbling,M.D. Squiers,Research EngineerCybernet Systems Corporationhttp://www.cybernet.comPhone: (313) 668-2567       (313) 668-8780 (FAX)------------------------------Date: Tue, 9 Apr 1996 13:45:00 -0400 (EDT)From: ben To: it-list@sparky.uthscsa.eduSubject: frame grabbersMessage-ID: A few questions:Is it possible to grab frames with a DT3155 running IT under windows 95?  I see that the Acquire menu has something for DTOL under windows NT but not 95.  What exactly is a TWAIN compliant device?- ben------------------------------Date: Thu, 11 Apr 1996 09:02:16 -0600 (CST)From: "S. Brent Dove" To: IT List Subject: Color ImagingMessage-ID: <01I3ERHNRM4U000HZI@uthscsa.edu>Hello IT users,We are presently making plans to add segmentation of image data based on color.  This is not a simpley task, there are many opinions as to what color space is the best to represent and manipulate for segmentation.In order to provide the best image analysis and processing tool on the planet we need your help.  We would like to know your opinion about how you might use color in either RGB, HSV or HSI to segment images.Please give your application and how you are presently using color as a discriminator or how you would like to.Thanks for your support.S. Brent Dove                            Voice: (210) 567-3333Diagnostic Sciences                      Fax:   (210) 567-3334University of Texas                      Email: dove@uthscsa.eduHealth Science Center                    Web:   ddsdx.uthscsa.eduSan Antonio, TX    USA                   ftp:   maxrad6.uthscsa.edu------------------------------Date: Thu, 11 Apr 1996 09:02:21 -0600 (CST)From: "S. Brent Dove" To: IT List Subject: Re: Fourier TransformMessage-ID: <01I3ERHRFYS8000HZI@uthscsa.edu>Hello IT users,Recently on the server>Has anybody experience in doing fourier-transforms and frequency-domain >filtering on images using ImageTool? Is there a pre-made tool box or >would the process involve programming?We can easily implement a fast fourier transform FFT and the inverse, but I would like to know from the users how they would use this capability in their work.  We have not implemented this because in our experience we have not had the need.  Please let us know what you would use this for and exactly what algorithms would need to be implemented to provide the necessary support for using FFT.Thanks for your support.S. Brent Dove                            Voice: (210) 567-3333Diagnostic Sciences                      Fax:   (210) 567-3334University of Texas                      Email: dove@uthscsa.eduHealth Science Center                    Web:   ddsdx.uthscsa.eduSan Antonio, TX    USA                   ftp:   maxrad6.uthscsa.edu------------------------------Date: Thu, 11 Apr 1996 09:35:23 -0500 (CDT)From: Hal Hawkins To: it-list@sparky.uthscsa.eduSubject: Re: Color ImagingMessage-ID: <01I3ESNPX5S8002NMF@BEACH.UTMB.EDU>The most elegant application of color analysis would be the ability to inputseparately as many as 8 different signals from different fluorescent dyesin the same section, obtained using narrow-band filters in a confocal microscope - and then to produce false-color representations of the datain any three colors.   Our current application is much simpler.  We want to be able to measure theamount of immunohistochemical reaction product in a counterstained section.Since we stain basal lamina, we would especially like to be able to takeline scans across the dark line of reaction product and convert the pointof minimum intensity to an absorbance value.  The reaction produce is brown on a blue background.  At times we also produce red or blue reactionproducts, also with a linear distribution. Many thanks for your hard work and for soliciting our input! Hal Hawkins, hal@utmb.edu------------------------------Date: Thu, 11 Apr 1996 21:12:37 +0100From: Alan Hale To: it-list@sparky.uthscsa.eduSubject: Re: Color ImagingMessage-ID: <316D67B5.1C18@celtic.co.uk>S. Brent Dove wrote:> Please give your application and how you are presently using color as a> discriminator or how you would like to.BrentThanks for the invitation.I would like to be able to identify and then analyse approximately circular objects on the basis of colour. In particular I am interested in doing this with photographs of lichen quadrats, where a quadrat may contain a number of lichen species, which it may be possible to distinguish on the basis of their colour (this will be easier with some combinations of species than others and perhaps only really applicable to crustose species). Any one species will have some degree of color variation, so it would be useful to be able to sample an area of colour within a lichen so as to indicate the degree of tolerance. ( I have seen other software where it is possible to do this, at least with approximately circular coloured blobs).Best of luck!Alan------------------------------Date: Fri Apr 12 17:29:25 1996From: gil@commtouch.co.il (gil)To: IT-LIST@sparky.uthscsa.eduSubject: how to add scaning to my app ?Message-ID: <19960412172925gil@gilhecht>if anyone done it before.... please email me:gil@commtouch.co.ilthanksGil.---------------------------------------------------Gil Hecht                       gil@commtouch.co.ilCommTouch Software, Inc.       Voice : 972-9-96344510,Technology Ave.             Fax   : 972-9-961053Ein-Vered,40696 Isreal.---------------------------------------------------------------------------------Date: Fri, 12 Apr 1996 11:35:20 -0400From: Gary Krause To: it-list@sparky.uthscsa.eduSubject: Re: Color ImagingMessage-ID: <1.5.4.16.19960412153520.1c3703ea@cms.cc.wayne.edu>I would like to echo this request.  We use multiple histologicstains and immunostains on a tissue slice and would like toquantitate each color and map them to false colors for publication.Thanks.GaryAt 09:39 AM 4/11/96 +0600, you wrote:>>The most elegant application of color analysis would be the abilityto input>separately as many as 8 different signals from differentfluorescent dyes>in the same section, obtained using narrow-band filters in a confocal>microscope - and then to produce false-color representations of thedata>in any three colors.>>Our current application is much simpler.  We want to be able tomeasure the>amount of immunohistochemical reaction product in a counterstainedsection.>Since we stain basal lamina, we would especially like to be able totake>line scans across the dark line of reaction product and convert thepoint>of minimum intensity to an absorbance value.  The reaction produce is>brown on a blue background.  At times we also produce red or bluereaction>products, also with a linear distribution.>>Many thanks for your hard work and for soliciting our input!>>Hal Hawkins, hal@utmb.edu>>>Gary S. KrauseDepartment of Emergency MedicineWayne State University, Detroit, MI. USA------------------------------Date: Tue, 16 Apr 1996 09:01:26 -0600 (CST)From: "S. Brent Dove" To: IT List Subject: ITSDK Zip fileMessage-ID: <01I3LQXCFG8U00037S@uthscsa.edu>Hello IT users,Several ImageTool developers have reported problems with unzipping the ITSDK.ZIP file.  This is because of the use of long file names.  The old DOS version of PKUNZIP does not handle this well.  An error message relating a corruption of the file is the usual response.  We have provided the latest version of WinZIP a shareware utility for Windows95 and Window NT.  This program is designed to handle long file names. The utility is located on our ftp server atftp://maxrad6.uthscsa.edu/ftp/pub/utils/winzip95.exeThis is a self extracting archive which will automatically setup WinZip on your machine.  Then use WinZip to unzip the ITSDK.ZIP.  You must be using a Windows95 machine to use WinZip. Be sure to register your copy of this great shareware program.Good luck.  With the advent of long files names it seems silly to continue to use things like ms_histo.ccp.  S. Brent Dove                            Voice: (210) 567-3333Diagnostic Sciences                      Fax:   (210) 567-3334University of Texas                      Email: dove@uthscsa.eduHealth Science Center                    Web:   ddsdx.uthscsa.eduSan Antonio, TX    USA                   ftp:   maxrad6.uthscsa.eduend of 			    IT-LIST Digest 17------------------------------