Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!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: <9009151558.AA02403@deimos.ads.com> Date: 15 Sep 90 16:50:58 GMT Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: Vision-List@ADS.COM Distribution: inet Organization: The Internet Lines: 282 Approved: vision-list@ads.com Vision-List Digest Sat Sep 15 08:50:58 PDT 90 - Send submissions to Vision-List@ADS.COM - Send requests for list membership to Vision-List-Request@ADS.COM Today's Topics: grouping Locating edges in a field of view Industrial Vision Metrology Conference VISION and NN; special issue of IJPRAI available document MAC AI demos ---------------------------------------------------------------------- Date: 11 Sep 90 14:57:28+0200 From: Tilo Messer Subject: grouping I am interested in grouping regions (not edges!) for increasing the performance of an object identification system. It is part of a planned real-time interpretation system of scenes taken from a moving camera. I found a few articles and papers about grouping of egdes (Lowe et. al.), but these don't fit. Is anybody else interested in this topic or does anybody know some theoretical and practical work in this area? I would be glad about some useful hints. Thanks, Tilo | |\ /| voice: ++ 49 89 48095 - 224 | | \/ | FORWISS, FG Cognitive Systems fax: ++ 49 89 48095 - 203 | | | Orleansstr. 34, D - 8000 Muenchen 80, Germany ------------------------------ Date: Wed, 12 Sep 90 11:05:03 EDT From: ICR - Mutual Group Subject: Locating edges in a field of view Here is an interesting real-world problem for you comp.ai.vision'aries out there: I have built a scanning unit which basically produces a picture in memory of a 2-D object (such as a peice of paper) passing under the scanning unit. The image is made only of a series of points outlining the object itself. The object passing under the scanner is roughly rectangular (i.e. four edges) but the edges can be somewhat bowed to make slightly concave or convex edges. There should be definate corners however. The problem is this. Given the limited information that I receive from the image, I must locate the edges of the object and calculate each side's length. The result should be a *very* accurate estimate of the height and width of the object and hence the area it covers. Oh ya, one other twist, the object can come through in any orientation. There is no guarantee a corner will always be first. Any ideas you have for algorithms, or documents you could point me toward would be greatedly appreciated! Like I said, an interesting problem. ------------------------------ Date: Tue, 11 Sep 90 11:05:00 EDT From: ELHAKIM@NRCCIT.NRC.CA Subject: Industrial Vision Metrology Conference ANNOUNCEMENT AND CALL FOR PAPERS INTERNATIONAL CONFERENCE ON INDUSTRIAL VISION METROLOGY Location: The Canadian Institute for Industrial Technology Winnipeg, Manitoba, Canada Date: July 11-13, 1991 Organized by: -International Society for Photogrammetry & Remote Sensing Commission V: Close-Range Photogrammetry and Machine Vision WG V/1: Digital and Real-time Close-range Photogrammetry Systems -National Research Council of Canada Proceeding published by: SPIE- The International Society for Optical Engineering Focusing on: Industrial applications of metric vision techniques Topics include: -Vision metrology techniques -Real-time systems -3-D object reconstruction -Decision algorithms -System calibration -Shop-floor metrology problems -Applications such as dimensional inspection 500-1000 words abstracts are to be submitted before January 1, 1991 to: Dr. S. El-Hakim National Research Council 435 Ellice Avenue Winnipeg, Manitoba, Canada R3B 1Y6 tel:(204) 983-5056 / Fax:(204) 983-3154 ------------------------------ Date: Thu, 13 Sep 90 15:29:49 PDT From: skrzypek@CS.UCLA.EDU (Dr. Josef Skrzypek) Subject: VISION and NN; special issue of IJPRAI Because of repeat enquiries about the special issue of IJPRAI (Intl. J. of Pattern Recognition and AI) I am posting the announcement again. IJPRAI CALL FOR PAPERS IJPRAI We are organizing a special issue of IJPRAI (Intl. Journal of Pattern Recognition and Artificial Intelligence) dedicated to the subject of neural networks in vision and pattern recognition. Papers will be refereed. The plan calls for the issue to be published in the fall of 1991. I would like to invite your participation. DEADLINE FOR SUBMISSION: 10th of December, 1990 VOLUME TITLE: Neural Networks in Vision and Pattern Recognition VOLUME GUEST EDITORS: Prof. Josef Skrzypek and Prof. Walter Karplus Department of Computer Science, 3532 BH UCLA Los Angeles CA 90024-1596 Email: skrzypek@cs.ucla.edu or karplus@cs.ucla.edu Tel: (213) 825 2381 Fax: (213) UCLA CSD DESCRIPTION The capabilities of neural architectures (supervised and unsupervised learning, feature detection and analysis through approximate pattern matching, categorization and self-organization, adaptation, soft constraints, and signal based processing) suggest new approaches to solving problems in vision, image processing and pattern recognition as applied to visual stimuli. The purpose of this special issue is to encourage further work and discussion in this area. The volume will include both invited and submitted peer-reviewed articles. We are seeking submissions from researchers in relevant fields, including, natural and artificial vision, scientific computing, artificial intelligence, psychology, image processing and pattern recognition. "We encourage submission of: 1) detailed presentations of models or supporting mechanisms, 2) formal theoretical analyses, 3) empirical and methodological studies. 4) critical reviews of neural networks applicability to various subfields of vision, image processing and pattern recognition. Submitted papers may be enthusiastic or critical on the applicability of neural networks to processing of visual information. The IJPRAI journal would like to encourage submissions from both , researchers engaged in analysis of biological systems such as modeling psychological/neurophysiological data using neural networks as well as from members of the engineering community who are synthesizing neural network models. The number of papers that can be included in this special issue will be limited. Therefore, some qualified papers may be encouraged for submission to the regular issues of IJPRAI. SUBMISSION PROCEDURE Submissions should be sent to Josef Skrzypek, by 12-10-1990. The suggested length is 20-22 double-spaced pages including figures, references, abstract and so on. Format details, etc. will be supplied on request. Authors are strongly encouraged to discuss ideas for possible submissions with the editors. The Journal is published by the World Scientific and was established in 1986. Thank you for your consideration. ------------------------------ Date: Wed, 5 Sep 90 13:21:08 +0200 From: ronse@prlb.philips.be Subject: available document The following unpublished working document is available. If you want a copy of it, please send me: - Your complete postal (snail mail) address, preferably formatted as on an enveloppe (cfr. mine below); an e-mail address is useless in this context. - The title of the working document. Christian Ronse Internet: ronse@prlb.philips.be BITNET: ronse%prlb.philips.be@cernvax Philips Research Laboratory Avenue Albert Einstein, 4 B-1348 Louvain-la-Neuve Belgium Tel: (32)(10) 470 611 (central) (32)(10) 470 637 (direct line) Fax: (32)(10) 470 699 ========================================================================= A twofold model of edge and feature detection C. Ronse September 1990 ABSTRACT. Horn's model of surface reflectance shows that edges in three-dimensional surfaces lead to grey-level edges combining in various ways sharp or rounded steps, lines and roofs. The perceptual analysis of extended edges necessicates the localization not only of step and line edges, but also of roof edges and Mach bands, and more generally of discontinuities and sharp changes in the n-th derivative of the grey-level. Arguments are given which indicate the inadequacy of locating features at zero-crossings of any type of smooth operator applied to the image, and the necessity of orientationally selective operators. The null space of feature detection is defined; it contains in particular all constant signals. Oriented local features are modelled as the linear superposition of a featureless signal (in the null space), an even-symmetric and/or an odd-symmetric feature, measured by convolution with respectively even-symmetric and odd-symmetric functions. Advantages of energy feature detectors are given. KEY WORDS. Edge types, zero-crossings and peaks, orientational selectivity, linear processing, feature symmetry, energy feature detector. ------------------------------ Date: Wed, 12 Sep 90 02:41:16 GMT From: pegah@pleiades.cps.msu.edu (Mahmoud Pegah) Subject: MAC AI demos Organization: Computer Science, Michigan State University, E. Lansing Greetings; I am trying to find freeware demos of AI that run on the MAC. This will be used in a classroom setting (not in a lab) and will be projected on a large screen from the video on the MAC. Demos having to do with search space techniques, natural language processing, vision, neural nets, knowledge based systems... would all be items I would like to FTP for use here. These demos will be used in an intro grad level survey course in AI. Reply to me directly, and indicate whether you would like your demo to be listed in a catalogue of AI educational demos that I will prepare from the mail I get. I will post the composed directory back to the net in two weeks time. Please indicate an FTP host (with internet number) from which your demo can be FTPed. Thanks in advance. -Mahmoud Pegah pegah@pleiades.cps.msu.edu AI/KBS Group pegah@MSUEGR.BITNET Comp Sci Dept ... uunet!frith!pegah Mich State Univ ------------------------------ End of VISION-LIST ********************