Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!ucbvax!ADS.COM!Vision-List-Request From: Vision-List-Request@ADS.COM (Vision-List moderator Phil Kahn) Newsgroups: comp.ai.vision Subject: Vision-List delayed redistribution Message-ID: <8907080500.AA08385@deimos.ads.com> Date: 7 Jul 89 22:15:02 GMT Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: Vision-List@ADS.COM Distribution: inet Organization: The Internet Lines: 169 Approved: vision-list@ads.com Vision-List Digest Fri Jul 07 14:15:02 PDT 89 - Send submissions to Vision-List@ADS.COM - Send requests for list membership to Vision-List-Request@ADS.COM Today's Topics: Vision-based solutions for the game of GO Summary of Image Processing Languages Re: Vision/Image Processing Languages E-mail addresses for range data ---------------------------------------------------------------------- Date: Thu, 29 Jun 89 17:43:25 -0500 From: uhr@cs.wisc.edu (Leonard Uhr) Subject: Re: Vision-based solutions for the game of GO Status: RO Al Zobrist used a vision-correlational approach to [the game of] GO in his Ph.D. thesis here (see diss abstracts). He published this in one of the Joint Computer Conferences (Eastern or Spring, around 1968-72). ------------------------------ Date: 29 Jun 89 19:14:47 GMT From: mdavcr!rdr@uunet.UU.NET (Randolph Roesler) Subject: Summary of Image Processing Languages Keywords: image vision language Organization: MacDonald Dettwiler, Richmond, B.C., Canada A couple of weeks ago I asked the net for pointers to image processing lanaguages (not libraries). I was looking for a computer programming language that would allow me to inter-image operations. That is, operations such as feature identifcation. It seems that such things don't really exist. All commercial packages that claimed "language" were really function libraries (choose your favorite language) with a little bit of control structure thrown in (a bonus ?). These libraries perform image to image transformations. They give you like access to the internals an image. Some systems were all control structure (KBVision as an example). These systems provided good end user environments, but not good for programmers or analyst tring to develop new image processing systems. (KBVision is great!, get it if you need an image processing system for "image" engineers, but its not really for programmers.) Further, most products were tied to specialized hardware. I work in the research department and have to live with stock SUN 3/60s. So even as libraries, most of these products are useless to us. I did receive three informed responces to my query. One person suggested a set of unix filters (that he wrote) as a useful image processing language. Another suggested IDL (image description langauge ?). The third forwarded seven good references on the subject. I am researching them now. Maybe, I'll write that image processing language myself. =====included================ I am not familiar with any image languages, per se, although in the past I've seen some references to them in the literature. I have written a general image processing software package (your image->image transformations, mostly) which consists of a large number of UNIX `filters' which can be connected up in relatively arbitrary ways using the UNIX `pipe' facility. If that sounds useful, let me know, and I'll mail you some blurbs. Mike Landy SharpImage Software P.O. Box 373, Prince St. Sta. New York, NY 10012-0007 (212) 998-7857 landy@nyu.nyu.edu =====included================ have you heard of idl on a sun/vax - pretty good, but very few high-level function in either space. call david stern research systems inc. (IDL) 303-399-1326 john j. bloomer =====included================ Take a look to Picasso and Pal: 1. Z. Kulpa. Picasso,, picasso-show and pal. a development of a high-level software system for image processing. pages 13-24 in (3). 2. T. Radhakrishnan, R. Barrera, et al. Design of a high level language (L) for image processing. pages 25-40 in (3). 3. Languages and Architectures for Image Processing. Academic Press, 1981. Editors: M. J. B. Duff, S. Levialdi. 4. MAC. Chapter four of (3). 5. A Language for parallel processing of arrays. Chapter five of (3). 6. PIXAL. Chapter sic of (3). 7. A high-level language for constructing image processing commands. Chapter 7 of (3). Also: Chapters Nine, ten, eleven, twelve, of (3). Adolfo Guzman. International Software Systems, Inc. 9420 Research Blvd., Suite 200. Austin, TX. 78759 Tel. (512) 338 1895 Telex: 499 1223 ISSIC Fax: (512) 338 9713 issi!guzman@cs.utexas.edu or guzman@issi.uucp or cs.utexas.edu!issi!guzman ========================== PS - my orginal posting never made it to Canada (or maybe, just not BC). So the responce may have been smaller than need be. ========================== Randy Roesler ...!uunet!van-bc!mdavcr!rdr MacDonald Dettwiler Up here in Canada. Image Processing Specialists. 604-278-3411 ------------------------------ Date: Fri, 30 Jun 89 12:16:29 GMT From: H Wang [Reply to Randy's original request for vision language information. pk...] Subject: Re: Vision/Image Processing Languages We developed an image-to-image language at Leeds on the transputer array, called APPLY, which was originally from CMU fro the Warp machine. APPLY performs localised window operations, eg. edge detection, 2D convolution. The APPLY compiler generates OCCAM (for transputer arrays), C (for UNIX machine) and W2 (for Warp) at the moment, although it is aimed at machine independent. This language has two major advantages: (1) ease the programming efforts, (2) efficiency. For instance, it does better on the Canny edge detector than the hand-crafted code reported in the literature. If you are interesting, pls contact me. I could not reach you by e-mail. My address is: Mr. H Wang, School of Comuter Studies, The university of Leeds, Leeds, LS2 9JT, phone: (0532) 335477 (in UK) +44532,335477 (international) e-mail: (ArpaNet)hw%uk.ac.leeds.dcs@uk.ac.ucl.cs.nss ------------------------------ Date: 3 Jul 89 11:54:00 WET From: John Illingworth Subject: e-mail addresses for range data hi, in recent vision lists there has been correspondence about range image data bases. I wish to obtain further information about these databases by email. However no email addresses have been given in the vision list. Anyone know the email address for M Rioux or L Cournoyer National Research Centre of Canada Ottawa, Ontario. Canada or Tom Henderson University of Utah many thanks John Illingworth ------------------------------ End of VISION-LIST ********************