Copy Link
Add to Bookmark
Report
VISION-LIST Digest Volume 10 Issue 41
VISION-LIST Digest Fri Oct 04 16:28:07 PDT 91 Volume 10 : Issue 41
- Send submissions to Vision-List@ADS.COM
- Vision List Digest available via COMP.AI.VISION newsgroup
- If you don't have access to COMP.AI.VISION, request list
membership to Vision-List-Request@ADS.COM
- Access Vision List Archives via anonymous ftp to ADS.COM
Today's Topics:
Vision List ftp has been flakey lately, but it is improving...
Remotely sensed images
Road Images
ftp-site for images and image-sequences.
Need the right hardware
Literature on Feature Grouping
Wanted: info on Image Processing Software systems
Feature detection in 1-dim. signals with non-stationary noise
"WANTED", a quad-tree based image segmentation algorithms
Brodatz Textures Revisited
Brodatz
Responses to Brodatz/Texture request
CAMP 91: Final Program Information & Deadline Reminder
Texture Images
Brodatz Textures
Texture images
Texture Images
textures
----------------------------------------------------------------------
Date: Thu, 3 Oct 91 09:29:33 -0700
From: vision@ads.com (Philip Kahn)
Subject: Vision List ftp has been flakey lately, but it is improving...
Our site has been upgrading our file system, and that has
temporarily impacted the reliability of our ftp access to the Vision
List Archives. If you experience a problem, first thing to do is
retry a short time later. If the problem persists, please notify me
at "vision-list-request@ads.com" and I'll check into the problem.
In the past, the ftp access has been very reliable, and I hope it will
soon be equally reliable. Sorry for any inconvenience.
phil...
------------------------------
Date: Tue, 1 Oct 91 16:34:22 CDT
From: hmueller@orca.tamu.edu (Hal Mueller)
Subject: Remotely sensed images
The distribution of the GIS Grass, available from uxc.cso.uiuc.edu
(anonymous, pub/grass), has some Spot and Landsat images of a region
in the north central US, together with vector and raster maps of the
same area.
Hal Mueller Tax the poor! Vote yes on the Texas lottery.
hmueller@orca.tamu.edu
------------------------------
Date: Tue, 1 Oct 91 09:42:21 -0400
From: harsh@eceris.ece.ncsu.edu (Harsh Potlapalli)
Subject: Road Images
We are looking for images of roads to be used for testing of a mobile
robot road follower, a la CMU's NAVLAB. Any pointers to where we
could locate such a database would be greatly appreciated.
Many thanks in advance.
Harsh
------------------------------
Date: 1 Oct 91 14:38:50 GMT
From: hobo@isy.liu.se (H}kan B}rman)
Subject: ftp-site for images and image-sequences.
Keywords: ftp images sequences
Organization: Dept of EE, University of Linkoping
We are installing a number of images, stereo pairs and image sequences
for ftp access. You get them by ftp to isy.liu.se (130.236.1.3) where
you login as ftp and change directory to:
/site/src/public/Packages/images
The purpose is to supply a broad selection of images and sequences for
testing and evaluation of different algorithms. This includes material
originating from ourselves (some have previously been distributed by
tape) as well as other sources.
The selection is not so broad yet (But I'm working on it :-). The
contents right now includes `Lenna', the `Swedish breakfast table'
stereo pair, and two ultrasound sequences of a heart.
All proposals and comments regarding the contents of this directory is
welcome.
/Hakan.
H}kan B}rman | Email: hobo@isy.liu.se
Linkoping University |
Dept. of Electrical Engineering |
Computer Vision Laboratory |
S-581 83 Linkoping | Phone: +13 28 18 85
Sweden | Fax: +13 13 85 26
------------------------------
Date: Fri, 4 Oct 91 11:28:46 -0700
From: marcel@saturn.ads.com (Marcel Schoppers)
Subject: need the right hardware
Can anyone tell me about some hardware that has the following properties:
REQUIRED CAPABILITIES
- the hardware digitizes 3 black-and-white cameras in real time and
(almost) simultaneously (e.g. by reading 3 black-and-white signals
as RGB color input);
- when used to digitize 3 black-and-white cameras, the digitizer
output should be three planes of light-intensity bytes;
- the hardware can be used as a frame grabber;
- video image resolution is at least 512 x 512 pixels;
- the hardware comes on one or more boards that are Sun-compatible
(either VMEbus or Sbus);
- the hardware can download parts/subregions of frames into RAM;
- while also functioning as frame grabber, the hardware can
simultaneously send each of the three video inputs straight
through to its own window on the Sun workstation's screen,
again without burdening the host processor;
- the hardware allows us to overlay line drawings onto the video
signal being displayed on the screen;
- it is reliable hardware, and in case of breakdown the vendor has
a quick replacement service (a few days);
- the software required to use the hardware can be written in one
labor week (or so);
- the hardware should cost much less than $20K.
NICE BUT OPTIONAL CAPABILITIES
- the hardware can subtract two non-consecutive frames on the
board, i.e. without burdening the host processor, and can
download the result to the Sun's RAM.
I have evaluated the DataCube MAXvideo20 board against this checklist, and
it satisfies all but two of my criteria. Its problems are that it will not
display the video on a Sun screen without hacking around inside the Sun's
hardware (so the only remaining option is to read the images into Sun memory,
which cuts the image redisplay rate to something like once per second), and
it costs around $20K. (Other assemblages of DataCube components with the
needed functionality will wind up costing about the same).
If anyone knows of hardware that meets my needs, please let me know ASAP,
and thanks very much.
Marcel Schoppers
------------------------------
Date: Tue, 1 Oct 91 17:42:54 +0100
From: Ma Ruihua <Ma.Ruihua@sophia.inria.fr>
Subject: Literature on Feature Grouping
I'm looking for references in the domain of feature grouping.
Features can be contour chains, line segments, curves, etc., both in
stereo images and a time-varying sequence of images. I'm rather a
beginner; so any infomation on this subject will be wellcome (review,
articles, ...)
Thanks a lot.
MA Ruihua
INRIA Centre de Sophia Antipolis / Project PASTIS
2004, Route des Lucioles / BP 109 / F-06561 Valbonne Cedex (France)
Tel: (33) 93.65.78.66 Email: ruihua@mirsa.inria.fr
------------------------------
Date: Tue, 1 Oct 91 16:30:18 EDT
From: dorai@cps.msu.edu (Chitra Dorai)
Subject: Wanted: info on Image Processing Software systems
Hi,
I'd appreciate any information on Image Processing Software
Systems (e.g. The HIPS Picture Processing software from Newyork
University) available on PCs, workstations, etc.
Specifically, I need information on machine vision oriented software
utilities and would appreciate any pointers, references regarding this.
Please reply via e-mail and if there is any interest, I`ll post
a summary.
Thanks,
Chitra Dorai.
Dept. Of Computer Science,
Michigan State University,
E.Lansing, MI 48823.
E-mail: dorai@pixel.cps.msu.edu
------------------------------
Date: Thu, 3 Oct 1991 09:38:16 GMT
From: waynet@kit.cna.tek.com (Wayne Turner;923-4517)
Subject: Feature detection in 1-dim. signals with non-stationary noise
I want to develop an algorithm for labelling features in a
1-dimensional signal with equally spaced samples. The signal contains
about 2000-16000 points and the waveform is composed of linear segments
with negative slope. The segments are seperated by "faults" which are
5-50 samples long (but fixed length for a given signal). After each
fault, a new segment begins with roughly the same slope as the previous
segment but with a drop in amplitude. In other words, "featureless"
segments of the waveform are simply lines with negative slope (with
noise added).
The noise is non-Gaussian, has low and high-frequency components and
the noise amplitude increases as the signal amplitude decreases.
Attempts to model the noise well enough to make traditional signal
processing techniques work have failed.
Can anyone point me to references or have ideas regarding this
application: specifically, techniques for feature and edge labelling
(and threshold determination) in 1-dim. signals with non-stationary
noise.
Thank you,
Wayne Turner
Tektronix, Inc.
waynet@kit.CNA.TEK.COM
------------------------------
Date: 3 Oct 91 09:08:06 GMT
From: yaonan@ph.tn.tudelft.nl (Yaonan Zhang)
Subject: "WANTED", a quad-tree based image segmentation algorithms
Summary: quad tree based image segmentation
Keywords: quad tree , image segmentation
Dear everybody,
I am looking for free algorithms of image segmentation based on quad tree or
other data structures, in C or Fortran.
Any pointers will be greatly appreciated!!
Yaonan
------------------------------
Date: Fri, 4 Oct 91 15:55:28 EDT
From: Stanley Dunn <smd@occlusal.rutgers.edu>
Subject: Brodatz Textures Revisited
Thanks to all of you for your interest in receiving the Brodatz textures.
I have discovered that there was a local network problem and
incisal.rutgers.edu was unreachable this past week. If you tried to get
the textures and failed, please try again as I think the problem has been
fixed. For those of you who did not request the information, I have taken
the liberty of reproducing it below.
THANK YOU for requesting information on access to
the texture archive. Hopefully, this is sufficient
information for you to gain access to the brodatz
textures and other ``natural'' textures in place on
the archive.
1. The texture samples are location on
incisal.rutgers.edu 128.6.45.13
and can be accessed by anonymous ftp.
2. There are two relevant directories in /usr1/u1/ftp
of interest: the first is the directory brodatz
and the second is the directory textures.
In the brodatz directory are the digitized images
of plates from the Brodatz album. As of this writing,
I have found 19 of my samples. As I get the rest of them
off of the magnetic tape, they will be added to the archive.
In the textures directory are other ``natural'' textures
that were created locally. Most of the file names are self-explanatory
and include, for example, kitty litter, rugs, walls, sponges, bread,
cat food and towels (I do not own a cat, but one of my PhD students
does).
3. The texture images in both directories are compressed and have
University of Maryland CVL format headers on them. To make it
easier for you to get to the raw images, I have included a UNIX
tar file containing the programs you will need to ``delete'' the
cvl headers (dh.c) and ``add'' cvl headers (ah.c). Sources of the
supporting subroutines are also included in the tar file. You should
only need to compile all the sources and link them together to generate
an executable program to delete the headers.
4. The samples in the brodatz directory are 256 x 256; those in the
textures directory are 512 x 512. In both cases they are 8 bits
per pixel.
5. If you encounter any difficulty in accessing the archives, please
contact me in one of the following ways and we can arrange an
alternative method for you to receive the texture samples. If you
encounter any problems in deleting the headers or with the images
themselves, then plese let me know. Needless to say, if everything
works fine, then I would like to know that too.
Best Regards,
Stan Dunn
smd@occlusal.rutgers.edu
(908) 932 4462 (Rutgers telephone)
(908) 932 3753 (Rutgers facsimile)
(201) 456 4777 (Medical school telephone)
------------------------------
Date: Fri, 04 Oct 91 11:16:26 -0700
From: "David A. Honig" <honig@aster.ICS.UCI.EDU>
Subject: Brodatz
>Date: Wed, 18 Sep 91 7:41:27 EDT
>From: Stanley Dunn <smd@occlusal.rutgers.edu>
>Subject: Brodatz Textures
>
>I have 23 of the plates from the Brodatz album digitized. These were
>digitized years ago when I was at Maryland and brought them with me
>to Rutgers. If there is enough interest in the textures, then I can
>move them to a machine in my lab where I have anonymous ftp set up
>and make them available to whoever is interested. Otherwise, I can
>just ftp them or mail to individuals who have requested them, as I
>have done in the past.
These plates were probably accompanied by a little (c) which
may attract the attention of people with little J.D.'s after their name...
------------------------------
Date: 2 Oct 91 10:47
From: chachere@cui.unige.ch
Subject: responses to Brodatz/Texture request
Several people asked if I could share the replies to my inquiry for
digitized images from Brodatz' Texture book:
*******************
Message inbox:20 - Read
From: <wen@computer-science.strathclyde.ac.uk>
Brodatz's book. I have some of the images. I can either e-mail to you or
you can send me a tape, then I put the images in and return the tape to your.
Wen Wen
*******************
From: Chiaofe Shu <chad@engin.umich.edu>
Subject: Texture images
Hi,
At the AI lab. of U. of Michigan, we have digitized Brodatz's texture book.
If you are still interested, I can set up some anonymous machine for
you to access these images.
Chiao-Fe
*******************
From: (Dane P. Kottke) <dkottke@ele.uri.edu>
Subject: Texture Images
Hi,
We've found a place to get digitized Brodatz Texture images. The
Digital Image Processing Lab at the University of Southern California
sells a set for $250 (us). The images are part of the Image Data Base
by Allen G. Weber. The 58 images come on 3, 9-track 1600 bpi tapes.
The person to contact is weber@willy.usc.edu
Apparantly, the images are not available via ftp. We hope to find a
site on the Internet, which will allow us to ftp the images at no
cost. If you know of a site, please reply with the information.
Thanks,
Dane Kottke
dkottke@ele.uri.edu
Dept. of Electrical Engineering,
Univeristy of Rhode Island
*******************
From: John Husoy-stip <jonh@tele.unit.no>
Subject: textures
Hi!
I don't know how to get hold of digitized images of all
of the images from brodatz book. I do, however, have a reasonable
amount of digitzed texture images some of which I believe are taken
from brodatz album. If you would like to check out what we have
I would be happy to give you a temporary user on our machine to enable
you to ftp some of the images.
john haakon husoy
norwegian inst. of technology
trondheim
norway
------------------------------
Date: Thu, 3 Oct 91 05:04:04 +0100
From: Fortunel Christian <fortunel@etca.fr>
Subject: CAMP 91: Final Program Information & Deadline Reminder
CAMP 91 in PARIS, FRANCE _____________
/ PARIS \
FINAL PROGRAM INFORMATION & DEADLINE REMINDER FOR REGISTRATION | _______ |
| / \ |
Title: Computer Architecture for Machine Perception | | | |
Date: 16, 17, 18 December 1991 | | | |
Place: PARIS, FRANCE (Hotel Forest Hill - La Villette) |__| |__|
SPONSORED by; DGA/DRET and MEN/DRED
CO-ORGANIZED by: CNRS/IEF and ETCA/CREA/SP
IN COOPERATION with: IEEE CIRCUITS & SYSTEMS SOCIETY and AFCET
General chair: B. Zavidovique
Program chair: P.L. Wendel
Program Committee: M. Bayoumi, R. Brodersen, V. Cantoni, P.E. Danielsson,
F. Devos, M. Ejiri, J. Gallice, J. Little, P. Matherat,
J. Sanz, S. Tanimoto
Local Arrangement: G. Stamon, P. Clermont (Treasurer), A. Merigot,
E. Pissaloux, R. Raynaud
*=======================================================================*
# REGISTRATION FEES #
# UNIVERSITY AFFILIATES' FEES #
# (Considering high living costs in Paris, we have obtained #
# a government grant to help decrease university affiliates #
# registration fees) #
# IEEE or AFCET Non members #
# ------------- ------------- #
# Before Nov 7th FF1300 - $250 FF1450 - $280 #
# After Nov 7th FF1600 - $300 FF1750 - $340 #
# #
# OTHERS' FEES #
# IEEE or AFCET Non members #
# ------------- ------------- #
# Before Nov 7th FF2150 - $405 FF2350 - $450 #
# After Nov 7th FF2600 - $485 FF2850 - $540 #
# #
# Registration covers #
# * One copy of hard-bound proceedings #
# * Open bar & buffet panel (Monday evening) #
# * Cruise dinner on the Seine river (Tuesday night) #
# * breafasts, lunches & coffe breaks during all 3 days #
# #
# ACCOMPANYING PERSONS' FEES (only meals as above & cruise) #
# #
# Before Nov 7th FF1340 - $270 #
# After Nov 7th FF1790 - $360 #
# #
# HOTEL #
# Hotel Forest Hill - La Villette #
# A block of rooms has been reserved at the hotel for attendees. #
# Prices are as follows: #
# * Single rooms: FF450 - $85 #
# * Double rooms: FF500 - $95 #
# #
# REGISTRATION & HOTEL ARRANGEMENTS #
# Must be made through: #
# SERVOTEL - 16 Sentier des Sablons - 94230 CACHAN - FRANCE #
# Ask for registration form or contact: #
# Tel: 33-1-47-40-86-00 #
# Fax: 33-1-46-63-40-53 #
# Telex:632 783 F #
# #
# FOR FURTHER INFORMATION #
# Email: zavido@etca.etca.fr #
# Phone: Sylvette / 33-1-42-31-97-21 (Scientific) #
# Servotel / 33-1-47-40-86-00 (Arrangements) #
# #
*=======================================================================*
SUNDAY, DECEMBER 15
Registration from 5PM to 7PM on site
MONDAY, DECEMBER 16
7.00: * Registration
8.30: * Welcome: B. Zavidovique
* Invited talk: "Analog VLSI for low level vision"
E. Vittoz, CSEM, Neuchatel, Switzerland
9.30 Session M1: VLSI
* A 65X76 VLSI Retina for Rough Vision
T. Bernard, P. NGuyen, F. Devos, B. Zavidovique
ETCA, PARIS XI (France)
* Image Functional Memory: Architecture and Performance
Y. Fujita - NEC (Japan)
10.20 Coffee break
10.50 Session M2: VLSI Based Architecture
* Real-Time Image Processing using Functional
Programming on a Dataflow Architecture
J. Serot, G. Quenot - ETCA (france)
* Towards Real-Time Haar Transforms from Pyramids to Pipelines
M.G. Albanesi, M. Ferretti - Pavia University (Italy)
* A Real-Time System for Image Processing
H. Waldburger, J.Y. Dufour, B. Kause - TTD (France)
* The SARNOFF Pyramid Chip
G. Van Der Wal - DSRC Princeton (USA)
12.30 Lunch break
14.00 Session A1: Programming Parallel Computer
* How to program and Configure a Heterogeneous Multiprocessor
M. Zeltner, B. Schnenwly, P. Shaeren, N. Guggenbuhl
SFIT, Zurich (Switzerland)
* A graphical Programming Environment for PAPIA2
A. Biancardi, V. Cantoni, U. Cei, M. Mosconi
Pavia University (Italy)
* A set-based Language for Prototyping Parallel Algorithms
R. Hummel, R. Kelly, S.F. Hummel - Courant Institute (USA)
* MAVIS: A Visual Shell for Computer Vision and Image Processing
T.J. Olson, N.G. Klop - Virginia University (USA)
* Vision Problems and Structured Parallelism
M. Furnarii, M. Bigliardo, M. Giordano, A. Massarotti
Arcofelice (Italy)
16.00 Coffee break
16.30 Session A2: Architectures for Symbolic Processing
* A Hardware Architecture for Tree Classifiers
P. Wanbacq, B. Dom, D. Steele - IBM San Jose (USA)
* Knowledge representations for Parallel Parsing Strategies
B. Habert - ENS Lyon (France)
* SNARS: A semantic Network Architecture for Rule-based Simulations
R.K. Karne, A.K. Sood - George Mason University (USA)
* Real-Time Multisensor Perception with the Experiental
Knowledge Base System
P.A. Ligomenides - Maryland University (USA)
18.15: Buffet/bar - Panel discussion
* AI on Massively Parallel machines
Moderator: S. Tanimoto
TIESDAY, DECEMBER 17
8.30: Invited talk
* Recent Progress on Massively Parallel Computing: routing
architectures and programming
J. Sanz IBM-PC (Argentina and USA)
9.30: Session M1: Communication Networks
* Multiscattering on a Reconfigurable Network of processors
J.J. Li - ENS Lyon (France)
* Compound Graph Networks for Parallel Image Processing
M. Hamdi, R.W. Hall - Pittsburgh University (USA)
10.20: Coffee break
Session M2: Heterogeneous Architectures
* Computational Architectures for Responsive Vision: The Vision Engine
J.J. Little, R. Barman, S. Kingdon, J.Lu
British Columbia University (Canada)
* A parallel Vision Machine: TRANSVISION
J.P. Derutin, B. Besserer, T. Tixier, A. Kikel
Clermont-Ferrand University (France)
* A Versatile Parallel Computer Architecture for Machine Vision:
Comparison of Real Images and CAD-Based Representations
E. Hirsch, P. Paillou, C. Muller, V. Gengenbach
LSIT/ENSP - Strasbourg (France)
* Concept and realization of a Heterogeneous Multiprocessor system
for Real Time Image Processing
A. Guzinger - SFIT Zurich (Switzerland)
12.30: Lunch break
14.00: Session A1: SIMD
* On the Convergence of a Pyramidal Euclidean Distance Transform Method
S.L. Tanimoto, P.E. Danielsson - Washington University (USA)
* Architectural Simulation of a Fine Grain Parallel Pyramid
Computer on the Connection Machine
E. Rougerie, A. Merigot - I.E.F. Paris XI University (France)
* Parallel Architectures for Multifocale Markov Random Field-Base
Image, Analysis
E. Memin, F. Charot, F. Heite - IRISA Rennes (France)
* Region-to-Region Visibility Analysis using Massively Parallel
Hypercube Machines
Y.A. Teng, D. Dementhon, L.S. Davis - Maryland University (USA)
* Parallel stereo and Image Matching on fixed size Processor Arrays
A. Khokmar, W.M. Lin, V. Prasanna
University of South California (USA)
16.00: Coffee Break
16.30: Session A2: Mathematical Morphology
* Cache Tiling for High Performance Morphological Image Processing
C.W. Wittenbrink, A.K. Somani - Washington University (USA)
* Speeding Up Mathematical Morphology Processes by Using a
New ASIC: PIMM1
R. Peyrard, J.C. Klein - ENSMP Fontainebleau (France)
* Fast Algorithms for Morphological Image Operations using
Bitmapped Binary Images
R. Van Der Boomgaard, R. Van Balen - Amsterdam University (Holland)
17.15 -19.15 Poster session
* Boundary closing with an asynchronous cellular automata network
J.F. Quesne, D. Demigny - ENSEA Cergy Pontoise (France)
* A Multilayer Perceptron Approach for Ornamental Stone Classification
G. Bonifazi, P. Burrascano - Roma University (Italy)
* A parallel Coprocessor Architecture for Binary Image Processing
J.D. Lecat, P. de Huelanaere - Catholic University of
Louvain (Belgique)
* An Architecture for a Tagged Production System
K.N. Kumar, V. Shridar, G. Krishna - Indian Institute of
Science - Bangalore (India)
* A Relaxation Computation of Optic Flow
J.F. Boyce, S.R. Protheroe, J.F. Haddon
Kings'College London (England)
* A Knoledge based Vision System for Recognizing 3D Objects
M.G. Milanova, A.K. Kunchen - Technical University Sofia (Bulgaria)
* Real-Time Movement Detection: The Stream Processor
F. Cabestaing, P. Bonnet, J.G. Postaire Lille University (France)
* Multiconstraints-Based Optical Flow Estimation and Segmentation
P. Nesi, A. Del Bimbo, J.L. Sanz - IBM (Argentina)
* The influence of Image Content and Architectural
Features on the Performance of Parallel Implementations
A.C. Daminanakis, S.C. Orphanoudakis - Institue of
Computer Science - Heraklion (Greece)
* Parallel Architecture for Image Processing with optic fiber
guided missile (Polypheme)
B. Longuet - Aerospatiale - Magny-les-Hameaux (France)
* Feature Extraction by the Partially Ranked Neuronal Layers
L. Bobrowski - Warsaw Institute of Biocybernetics and
Biomedical Engineering (Poland)
* Parallel Processing of High Level Symbolic Computer Vision Algorithms
C. Reinhart, R. Nevatia - University of South California
Los Angeles (USA)
* Parallel Processing of Image Contours
L.T. Chen - University of Maryland (USA)
* Constant Time Image Component Labelling on a Reconfigurable
Processor Array
H.M. Alnuweiri - Vancouver (Canada)
* Multiprocessor-based Simulated Annealing for Ellipse
and Parallelogram Detection
G. Bongiovanni, P. Crescenzi, C. Guerra - Roma University (Italie)
20.00: Cruise on the Seine River
WEDNESDAY, DECEMBER 18
8.30: Invited talk
* Optics in Computing Machines suitable for 2D-Data
P. Chavel - IOTA University Paris XI (France)
9.30: Session M1: Optical Computing
* Analog Optoelectronic Neuron Unit for Boltzmann Machines
P. Lalanne, J.C. Rodier, J. Taboury - IOTA Paris XI Univ. (France)
* D-Stop: An Optoelectronic Generalized Matrix Algebra
Parallel Processor
G.C. Marden, A.V. Kridshnamourthy, J. Merckle, G. Yayla,
J.E. Ford, S.L. Esener California University San Diego (USA)
10.20: Coffee Break
10.50: Session M2: Robot Vision Architecture
* A VLSI Architecture for Stereo Image Sequence Analysis
G. Seethbraman, M. Bayoumi, K. Valavanis M.C. Mulder
South Louisiana University, Lafayette (USA)
* Toward a Dedicated Architecture for Obstacle Detection and Avoidance
T. Maurin, R. Reynaud, B. Berschandy, A. Chebira
IEF - Paris XI University (France)
* Obstacle Detection using Bi-Spectrum CCD Camera and Image Processing
H.G. NGuyen, J.Y. Laisne - Renault Research Center
Reuil Malmaison (France)
* Parallel Architecture for Visual Servicing Applications
P. Martinet, P. Rives, P. Fickinger, J.J. Borrelly
Laboratoire d'Electronique - Aubiere (France)
12.30: Lunch break
14.00: Session A1: Application Oriented Systems
* Vision Processor for Moving Object Analysis
H. Kubota, Y. Okamoto, H. Mizoguchi, Y. Kuno
Toshiba Corporation - Kawasaki (Japan)
* A Fast and Intelligent Character Recognition System
Integrating Image-type and Logical-type Processings
H. Aso, S. Ohmachi, N. Sun, Y. Mizoguchi)
* Automated High Speed Printing Plate Checker
T. Agni, M. Tsuzuki, N. Nagamahi, T. Carew, K. Nakauchi
Tokyo Institute of Technology - Yokohama (Japan)
* An Optimal Autoregressive Identification Method for Industrial
Object Recognition
D. Ionescu - University of Ottawa - (Canada)
15.40 Coffee Break
16.00 Session A2: Architectural Comparisons
* Region Labelling in Image Processing with SIMD and MIMD
Architectures
P. Fernandez, D. Houzet, J.L. Basille - IRIT Toulouse (France)
* Performances Evaluation of some Divergent Parallel Architecture in
the Context of Filtering Algorithm using the CFRD Method in Image
R. Dapolguy, A. Dion - IUT Le Creusot (France)
17.00 Concluding panel discussion
* Designing yet other specialized machines
Moderator: T. Van der Pyl - CEC
------------------------------
End of VISION-LIST digest 10.41
************************
