Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!usc!ucsd!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: <8912070500.AA08613@deimos.ads.com> Date: 6 Dec 89 18:34:59 GMT Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: Vision-List@ADS.COM Distribution: inet Organization: The Internet Lines: 653 Approved: vision-list@ads.com Vision-List Digest Wed Dec 06 10:34:59 PDT 89 - Send submissions to Vision-List@ADS.COM - Send requests for list membership to Vision-List-Request@ADS.COM Today's Topics: We need "real world" images. Program For Realistic Images Wanted MATLAB Accuracy measure RE:Boundary tracking algorithms.. Computer Vision Position Post-Doctoral Research Fellowship New Journal: Journal of Visual Communication and Image Representation Boundary Tracking -----> (collected Information). Bibliography: 3D Scene Interp. Workshop (Austin, TX) ---------------------------------------------------------------------- Date: 27 Nov 89 19:48:16 GMT From: rasure@borris.unm.edu (John Rasure) Subject: We need "real world" images. Organization: University of New Mexico, Albuquerque We need images. Specifically, we need stereo pair images, fly by image sequences, LANDSAT images, medical images, industrial inspection images, astronomy images, images from lasers, images from interferometers, etc. The best images are those that correspond to a "typical" image processing problems of today. They need not be pleasing to look at, just representative of the imaging technique and sensor that is being used. Does anybody have such images that they can share? John Rasure rasure@bullwinkle.unm.edu Dr. John Rasure Department of EECE University of New Mexico Albuquerque, NM 87131 505-277-1351 NET-ADDRESS: rasure@bullwinkle.unm.edu ------------------------------ Date: Wed, 29 Nov 89 11:25:37 +0100 From: sro@steks.oulu.fi (Sakari Roininen) Subject: Program For Realistic Images Wanted We are preparing research project in the field of visual inspection. In our research work we want to compute and simulate highly realistic images. Key words are: Shading - Illumination We are looking for a software package including following properties (modules): - Geometric description of objects. - Optical properties of the surfaces. Surfaces of interest are: metal, wood, textile. - Physical and geometrical description of the light sources. - Physical and technological properties of the cameras. Software should be written in C and source code should be available so that we can customize the software to fit our applications. GOOD IDEAS ARE ALWAYS WELCOME !!! Please, contact: Researcher Timo Piironen Technical Research Centre of Finland Electronics Laboratory P.O.Box 200 SF-90571 OULU Finland tel. +358 81 509111 Internet: thp@vttko1.vtt.fi ------------------------------ Date: 1 Dec 89 18:00:26 GMT From: Adi Pranata Subject: MATLAB Hi, I'm not sure where to posted this question, anyway Does any one have any info, on convert raster images/picture to matlab matrix format, since i am interested on use the matlab software to manipulate it . Since it will be no problem to display the matlab file format using the imagetool software. Any info including what other newsgroup more appropriate to posted will be welcome. Thanks in advance. You could reply to pranata@udel.edu Sincerely, Desiderius Adi Pranata PS: Electromagnetig way 146.955 Mhz -600 KHz Oldfashioned way (302)- 733 - 0990 (302)- 451 - 6992 [ This is definitely appropriate for the Vision List. Answers to the List please. phil...] ------------------------------ Date: 2 Dec 89 17:41:18 GMT From: muttiah@cs.purdue.edu (Ranjan Samuel Muttiah) Subject: Accuracy measure Organization: Department of Computer Science, Purdue University I am looking for the various ACCURACY measures that are used in the vision field. If you have any information on this, could you email/post please ? Thank you. ------------------------------ Date: Wed, 29 Nov 89 18:47:24 EST Subject: RE:Boundary tracking algorithms.. From: Sridhar Ramachandran I have pseudo code for a Boundary Tracking Algorithm for Binary Images that uses Direction codes and Containment codes to track the boundary. It is pretty efficient and works fine. If interested, please e-mail requests to sramacha@uceng.uc.edu (OR) sridhar@uc.edu (OR) sramacha@uc.edu. Sridhar Ramachandran. ------------------------------ Date: Tue, 5 Dec 89 14:08:30 EST From: peleg@grumpy.sarnoff.com (Shmuel Peleg x 2284) Subject: Computer Vision Position - David Sarnoff Research Center The computer vision research group at David Sarnoff Research Center has an opening for a specialist in image processing or computer vision who has an interest in computer architecture and digital hardware. Master's level or equivalent experience is preferred. This is a key position in an established research team devoted to the development of high performance, real-time vision systems. The group is active at all levels of research and development from basic research to applications and prototype implementation. Current programs include object recognition, motion analysis, and advanced architecture. Please send your vitae or enquire with Peter Burt (Group Head), David Sarnoff Research Center, Princeton, NJ 08543-5300; E-Mail: burt@vision.sarnoff.com. ------------------------------ Date: Wed, 29 Nov 89 19:11:55 WET DST From: "D.H. Foster" Subject: Post-Doctoral Research Fellowship UNIVERSITY OF KEELE Department of Communication & Neuroscience POST-DOCTORAL RESEARCH FELLOWSHIP Applications are invited for a 3-year appointment as a post-doctoral research fellow to work in the Theoretical and Applied Vision Sciences Group. The project will investigate early visual form processing, and will entail a mixture of computational modelling and psychophysical experiment. The project is part of an established programme of research into visual information processing, involving a team of about ten members working in several well-equipped laboratories with a range of computing and very high resolution graphics display systems. Candidates should preferably be experienced in computational vision research, but those with training in computing science, physics, experimental psychology, and allied disciplines are also encouraged to apply. The appointment, beginning 1 January 1990, or soon thereafter, will be on the Research IA scale, initially up to Point 5, depending on age and experience. Informal enquiries and applications with CV and the names of two referees to Prof David H. Foster, Department of Communication & Neuroscience, University of Keele, Keele, Staffordshire ST5 5BG, England (Tel. 0782 621111, Ext 3247; e-mail D.H.Foster@uk.ac.keele). ------------------------------ Date: Thu, 23 Nov 89 16:48:13 EST From: zeevi@caip.rutgers.edu (Joshua Y. Zeevi) Subject: New Journal: Journal of Visual Communication and Image Representation New Journal published by Academic Press --------------------------------------- Dear Colleague, The first issue of the Journal of Visual Communication and Image Representation is scheduled to appear in September 1990. Since the journal will cover topics in your area of expertise, your contribution will most likely have impact on future advancements in this rapidly developing field. Should you have questions regarding the suitability of a specific paper or topic, please get in touch with Russell Hsing or with me. The deadline for submission of papers for the first issue is Feb. 15, and for the second issue May 15, 1990. For manuscript submission and/or subscirption information please write or call Academic press, Inc. 1250 6th Ave., San Diego, CA 92101. (619) 699-6742. Enclosed please find the Aims & Scope (including list of preferred topics) and list of members of the the Editorial Board. JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION -------------------------------------------------------- Dr. T. Russell Hsing, co-editor, Research Manager, Loop Transmission & Application District, Bell Communication Research, 445 South Street, Morristown, NJ 07960-1910 (trh@thumper.bellcore.com (201) 829-4950) Professor Yehoshua Y. Zeevi*, co-editor, Barbara and Norman Seiden Chair, Department of Electrical Engineering, Technion - Israel Ins. of Technology, Haifa, 32 000, Israel * Present address: CAIP Center, Rutgers University, P. O. Box 1390, Piscataway, NJ 08855-1390 (zeevi@caip.rutgers.edu (201) 932-5551) AIMS & SCOPE The Journal of Visual Communication and Image Representation is an archival peer-reviewed technical journal, published quarterly. With the growing availability of optical fiber links, advances of large scale integration, new telecommunication services, VLSI-based circuits and computational systems, as well as the rapid advances in vision research and image understanding, the field of visual communication and image representation will undoubtedly continue to grow. The aim of this journal is to combine reports on the state- of-the-art of visual communication and image representation with emphasis on novel ideas and theoretical work in this multidisciplinary area of pure and pure and applied research. The journal consists of regular papers and research reports describing either original research results or novel technologies. The field of visual communication and image representation is considered in its broadest sense and covers digital and analog aspects, as well as processing and communication in biological visual systems. Specific areas of interest include, but are not limited to, all aspects of: * Image scanning, sampling and tessellation * Image representation by partial information * Local and global schemes of image representation * Analog and digital image processing * Fractals and mathematical morphology * Image understanding and scene analysis * Deterministic and stochastic image modelling * Visual data reduction and compression * Image coding and video communication * Biological and medical imaging * Early processing in biological visual systems * Psychophysical analysis of visual perception * Astronomical and geophysical imaging * Visualization of nonlinear natural phenomena Editorial Board R. Ansari, Bell Communications Research, USA I. Bar-David, Technion - Israel Institute of Technology, Israel R. Bracewell, Stanford University, USA R. Brammer, The Analytic Sciences Corporation, USA J.-O. Eklundh, Royal Institute of Technology, Sweden H. Freeman, Rutgers University, USA D. Glaser, University of California at Berkeley, USA B. Julesz, Caltech and Rutgers University, USA B. Mandelbrot, IBM Thomas J. Watson Research Center, USA P. Maragos, Harvard University, USA H.-H Nagel, Fraunhofer-Institut fur Informations-und Datenverbeitung, FRG A. Netravali, AT&T Bell Labs, USA D. Pearson, University of Essex, England A. Rosenfeld, University of Maryland, USA Y. Sakai, Tokyo Institute of Technology, Japan J. Sanz, IBM Almaden Research Center, USA W. Schreiber, Massachusetts Institute of Technology, USA J. Serra, Ecole Nationale Superieure des Mines de Paris, France M. Takagi, University of Tokyo, Japan M. Teich, Columbia University, USA T. Tsuda, Fujitsu Laboratories Ltd., Japan S. Ullman, Massachusetts institute of technology, USA H. Yamamoto, KDD Ltd., Japan Y. Yasuda, University of Tokyo, Japan ------------------------------ Date: Mon, 27 Nov 89 16:48 EDT From: SRIDHAR BALAJI Subject: Boundary Tracking -----> (collected Information). X-Vms-To: IN%"Vision-List@ADS.COM" Status: RO /** This are some refs. and psedocode for the Boundary tracking I asked. ** Thanks so much for all the contributors. Since so many ** wanted this. I thought it may be useful to send it to the group. S. Balaji */ ******* From: IN%"marra@airob.mmc.COM" 14-NOV-1989 15:28:27.22 CC: Subj: Re: Boundary tracking. Here is Pascal-pseudo code for our binary boundary tracker. A minor mod will extend it to handle multiple objects. Good luck. Pseudo Code for the ALV border tracing module Program pevdcrok ( input,output ); (* include csc.si TYPE and VAR declarations *) (* this causes declaration of the following data elements: dir_cue version V_IM O_IM PB *) (* include peve.si TYPE and VAR declarations the following are defined in peve.si: pevdcrok_static_block.dc pevdcrok_static_block.dc *) (* ____________________FORWARD DECLARATIONS____________________ *) (* ----------OURS---------- *) (* ----------THEIRS-------- *) procedure pevdcrok(road,obst:imagenum) TYPE border_type_type = (blob,bubble) direction_type = (north,south,east,west) VAR inside_pix : d2_point; (* Col,row location of a pixel on the inside of a border *) outside_pix : d2_point; (* Col,row location of a pixel on the outside of a border *) next_pix : d2_point; (* Col,row location of the next pixel to be encountered during the tracing of the border *) next_8_pix : d2_point; (* Col,row location of the next eight-neighbor pixel to be encountered during the tracing of the border *) westmost_pix : d2_point; (* col,row location of the westmost pixel visited so far this image *) eastmost_pix : d2_point; (* col,row location of the eastmost pixel visited so far this image *) northmost_pix : d2_point; (* col,row location of the northmost pixel visited so far this image *) southmost_pix : d2_point; (* col,row location of the southmost pixel visited so far this image *) border_type : border_type_type; (* a processing control flag indicating the type of border assumed to be following *) direction : direction_type; (* directions being searched *) start_time, end_time, print_time : integer; (* recorded times for time debug *) procedure find_border(inside_pix,outside_pix,direction) begin (* find_border *) Set a starting point for finding blob in the middle of the bottom of the image if PB.D[pevdcrok,gra] then mark the starting point Search in direction looking for some blob, being sure you don't go off the top of the road image Search in direction looking for a blob/non-blob boundary, being sure you don't go off the top of the blob image if PB.D[pevdcrok,gra] then mark the inside_pix and the outside_pix end (* find_border *) procedure trace_border(border_type,inside_pix,outside_pix,direction) TYPE dir_type = (0..7); (* 0 = east 1 = northeast 2 = north 3 = northwest 4 = west 5 = southwest 6 = south 7 = southeast *) VAR dir : dir_type; (* relative orientation of the inside_pix outside_pix 2-tuple *) begin (* trace_border *) remember the starting inside and outside pix's for bubble detection set dir according to direction while we haven't found the end of this border do begin (* follow border *) next_pix.col := outside_pix.col + dc[dir]; next_pix.row := outside_pix.row + dr[dir]; while road.im_ptr^[next_pix.col,next_pix.row] = 0 do begin (* move the outside pixel clockwise *) outside_pix = next_pix; advance the dir check for bubbles; if a bubble then begin (* bubble has been found *) border_type := bubble exit trace_border end (* bubble has been found *) next_pix.col := outside_pix.col + dc[dir]; next_pix.row := outside_pix.row + dr[dir]; end (* move the outside pixel clockwise *) update the direction for moving inside_pix next_pix.col := inside_pix.col + dc[dir]; next_pix.row := inside_pix.row + dr[dir]; next_8_pix.col := inside_pix.col + dc[dir]; next_8_pix.row := inside_pix.row + dr[dir]; while road.im_ptr^[next_pix.col,next_pix.row] = 0 or (road.im_ptr^[next_8_pix.col,next_8_pix.row] = 0 and mod(dir,2) = 0) do begin (* move the inside pixel counter-clockwise *) inside_pix := next_pix; advance the dir if road.im_ptr^[inside_pixel.col,inside_pixel.row] <> 0 then begin inside_pix := next_8_pix; advance the dir; end; check for bubbles; if a bubble then begin (* bubble has been found *) border_type := bubble exit trace_border end (* bubble has been found *) next_pix.col := outside_pix.col + dc[dir]; next_pix.row := outside_pix.row + dr[dir]; end (* move the inside pixel counter-clockwise *) update the direction for moving outside_pix update values of westmost_pix,eastmost_pix,northmost_pix, southmost_pix if mod(num_border_points,crock_rec.boundary_skip) = 1 then record column and row values in V_IM.edge_record end (* follow border *) border_type := blob end (* trace_border *) begin (* pevdcrok *) if PB.D[pevdcrok,time] then clock(start_time); AND the road image with the border image, leaving the result in road image border_type := bubble while border_type = bubble do begin if PB.D[pevdcrok,tty] writeln('PEVDCROK: Calling find_border'); find_border(inside_pix,outside_pix,west) initialize IM edge_record; num_border_points := 0 if PB.D[pevdcrok,tty] writeln('PEVDCROK: Calling trace_border'); trace_border(border_type,inside_pix,outside_pix,west) end complete IM edge_record if PB.D[pevdcrok,time] then begin (* time debug *) clock(end_time); print_time := end_time - start_time; writeln('PEVDCROK: elapsed time = ',print_time,' msec'); end; (* time debug *) end (* pevdcrok *) ******* From: IN%"mv10801@uc.msc.umn.edu" 14-NOV-1989 16:04:34.13 CC: Subj: Re: Motion tracking See: J.A.Marshall, Self-Organizing Neural Networks for Perception of Visual Motion, to appear in Neural Networks, January 1990. ******* From: IN%"pell@isy.liu.se" "P{r Emanuelsson" 15-NOV-1989 13:24:37.17 CC: Subj: Re: Boundary tracking. I think you want to do chain-coding. The best method I know was invented by my professor (of course...) and is called "crack coding". It uses a two-bit code and avoids backtracking problems and such. It's quite easy to implement, but I don't think I have any code handy. The algorithm is, however, given as flow charts in the second reference: "Encoding of binary images by raster-chain-coding of cracks", Per-Erik Danielsson, Proc. of the 6th int. conf. on Pattern Recognition, Oct. -82. "An improved segmentation and coding algorithm for binary and nonbinary images", Per-Erik Danielsson, IBM Journal of research and development, v. 26, n. 6, Nov -82. If you are working on parallel computers, there are other more suitable algorithms. Please summarize your answers to the vision list. Cheers, /Pell Dept. of Electrical Engineering pell@isy.liu.se University of Linkoping, Sweden ...!uunet!isy.liu.se!pell ------------------------------ Date: Thu, 30 Nov 89 13:20:28 EST From: flynn@pixel.cps.msu.edu Subject: Bibliography: 3D Scene Interp. Workshop (Austin, TX) Here's a list of papers in the proceedings of the IEEE Workshop on Interpretation of 3D Scenes held in Austin, Texas on November 27-29. STEREO ------ R.P. Wildes, An Analysis of Stereo Disparity for the Recovery of Three-Dimensional Scene Geometry, pp. 2-8. S. Das and N. Ahuja, Integrating Multiresolution Image Acquisition and Coarse-to-Fine Surface Reconstruction from Stereo, pp. 9-15. S.D. Cochran and G. Medioni, Accurate Surface Description from Binocular Stereo, pp. 16-23. SHAPE FROM X ------------ R. Vaillant and O.D. Faugeras, Using Occluding Contours for Recovering Shape Properties of Objects, pp. 26-32. P.K. Allen and P. Michelman, Acquisition and Interpretation of 3D Sensor Data from Touch, pp. 33-40. P. Belluta, G. Collini, A. Verri, and V. Torre, 3D Visual Information from Vanishing Points, pp. 41-49. RECOGNITION ----------- R. Kumar and A. Hanson, Robust Estimation of Camera Location and Orientation from Noisy Data Having Outliers, pp. 52-60. J. Ponce and D.J. Kriegman, On Recognizing and Positioning Curved 3D Objects from Image Contours, pp. 61-67. R. Bergevin and M.D. Levine, Generic Object Recogfnition: Building Coarse 3D Descriptions from Line Drawings, pp. 68-74. S. Lee and H.S. Hahn, Object Recognition and Localization Using Optical Proximity Sensor System: Polyhedral Case, pp. 75-81. MOTION ------ Y.F. Wang and A. Pandey, Interpretation of 3D Structure and Motion Using Structured Lighting, pp. 84-90. M. Xie and P. Rives, Towards Dynamic Vision, pp. 91-99. ASPECT GRAPHS ------------- D. Eggert and K. Bowyer, Computing the Orthographic Projection Aspect Graph of Solids of Revolution, pp. 102-108. T. Sripradisvarakul and R. Jain, Generating Aspect Graphs for Curved Objects, pp. 109-115. D.J. Kriegman and J. Ponce, Computing Exact Aspect Graphs of Curved Objects: Solids of Revolution, pp. 116-121. SURFACE RECONSTRUCTION ---------------------- C.I. Connolly and J.R. Stenstrom, 3D Scene Reconstruction from Multiple Intensity Imagesm pp. 124-130. R.L. Stevenson and E.J. Delp, Invariant Reconstruction of Visual Surfaces, pp. 131-137. P.G. Mulgaonkar, C.K. Cowan, and J. DeCurtins, Scene Description Using Range Data, pp. 138-144. C. Brown, Kinematic and 3D Motion Prediction for Gaze Control, pp. 145-151. 3D SENSING ---------- M. Rioux, F. Blais, J.-A. Beraldin, and P. Boulanger, Range Imaging Sensors Development at NRC Laboratories, pp. 154-159. REPRESENTATIONS --------------- A. Gupta, L. Bogoni, and R. Bajcsy, Quantitative and Qualitative Measures for the Evaluation of the Superquadric Models, pp. 162-169. F.P. Ferrie, J. Lagarde, and P. Whaite, Darboux Frames, Snakes, and Super-Quadrics: Geometry from the Bottom-Up, pp. 170-176. H. Lu, L.G. Shapiro, and O.I. Camps, A Relational Pyramid Approach to View Class Determination, pp. 177-183. APPLICATIONS ------------ I.J. Mulligan, A.K. Mackworth, and P.D. Lawrence, A Model-Based Vision System for Manipulator Position Sensing, pp. 186-193. J.Y. Cartoux, J.T. Lapreste, and M. Richetin, Face Authentification or Recognition by Profile Extraction from Range Images, pp. 194-199. J.J. Rodriguez and J.K. Aggarwal, Navigation Using Image Sequence Analysis and 3-D Terrain Matching, pp. 200-207. ------------------------------ End of VISION-LIST ********************