Copy Link
Add to Bookmark
Report
VISION-LIST Digest 1990 02 20
Vision-List Digest Tue Feb 20 17:40:23 PDT 90
- Send submissions to Vision-List@ADS.COM
- Send requests for list membership to Vision-List-Request@ADS.COM
Today's Topics:
Others ways to do triangulation sought
Request for advice on equipment
Range data archive
Currently available packages for image processing
digital photography
CFP: IEEE TPAMI Special Issue on 3D Scene Interpretation
CVGIP TOC, Vol. 50, No. 1, March 1990
Conference on Visual Information Assimilation in Man and Machine
VBC - 90 Preliminary Conference Announcement
----------------------------------------------------------------------
Date: 19 Feb 1990 11:52:39-GMT
From: aa538 <aa538%city.ac.uk@NSFnet-Relay.AC.UK>
Subject: Others ways to do triangulation sought
I am developing a representation for image structure which involves
triangulating a set of (mostly) irregularly spaced data points. The
heuristic triangulation algorithm I developed is quite fast, but regularly
makes mistakes. I would be grateful if anyone could provide code (preferably C)
to perform the triangulation more robustly. Delauny triangulation is the only
type I know, but any one would probably be fine.
Paul Rosin
Machine Vision Group
Centre for Information Engineering
Dept. Electronic, Electrical, and Information Engineering
City University
Northampton Square
London, ECIV OHB
------------------------------
Date: Sun, 18 Feb 90 23:18:12 EST
From: yehuda@acsu.buffalo.edu (yehuda newberger)
Subject: request for advice on equipment
I need information on what kind of monitor and video card would be
appropriate for work in image analysis on MSDOS type equipment. I have
a 386 running MSDOS. Typically, I want to display 256 by 256 pixels
with 256 different simultaneous shades of gray or colors. I would prefer
to spend less than $1000 . My address is Edward Newberger
90 Huntington Avenue
Apt 104
Buffalo, New York 14214
------------------------------
Date: Mon, 19 Feb 90 11:45:59 EST
From: flynn@pixel.cps.msu.edu (Patrick J. Flynn)
Subject: Range data archive
I've made 44 range images (obtained from our Technical Arts
scanner) available for anonymous ftp from:
styrofoam.cps.msu.edu (IP address 35.8.56.144)
in the pub/images directory.
Some images contain one object, some contain several, with various
amounts of occlusion.
Direct *specific* questions about the images to me (flynn@cps.msu.edu).
General questions about range sensing are best answered by reading
the surveys by Jarvis (PAMI '83), Nitzan (PAMI '88), or Besl
(in the `Advances in Machine Vision' book by J. Sanz, pub. by Springer).
Here is the text of the README file in the images directory.
This directory contains a bunch of range images produced by the
MSU Pattern Recognition and Image Processing Lab's Technical
Arts 100X scanner (aka `White scanner'). You are free to use
these images to test your algorithms. If the images are to appear
in a published article, please acknowledge the MSU PRIP Lab as
the source of the images (you don't have to mention my name, though).
File format: rather than deal with all the goofy standards out
there for images (and to preserve the floating-point representation),
these images are compressed ASCII text files. Beware: they expand by
about 10x when uncompressed. I recommend that you keep them
compressed to save disk space. Many of you will probably convert
these files to your own `local' image format anyway.
Each image file has a three-line header giving the number of rows and
columns. This is followed by four images. The first is the
so-called 'flag' image, where a pixel value of 1 means the corresponding
(x,y,z) values at that pixel are valid. If the flag value is zero, you
should ignore the (x,y,z) components for that pixel.
Following the flag image is the image of X-coordinates, the image
of Y-coordinates, and the image of Z-coordinates. All are
floating-point images. Our White scanner is configured so that
each stripe of range values occupies one column in the image. We
sweep the object under the stripe with an XY table to get an image.
So the X coordinate image is a linear ramp; the X value is taken
from the absolute position of the X stage in the XY table (we don't
do anything in the Y direction at present). The Y value depends
on the column number of the pixel, and the Z value is the measured
range (in our lab, Z is the height above a table).
You can use the 3D coordinates of each range pixel, or you can
throw away the X and Y images, and concern yourself with the Z-value
alone. Note that the `aspect ratio' of the image doesn't
have to be 1, although I try to keep it in the neighborhood of 1.
Availability: I will try to keep these images available on
styrofoam.cps.msu.edu (35.8.56.144) until I leave MSU this summer.
If my next job has machines with internet access and some disk space,
I'll put them there.
Remember to use binary mode when you transfer the images.
------------------------------
Date: Wed, 14 Feb 90 17:25:43 PST
From: Scott E. Johnston <johnston@zooks.ads.com>
Subject: Currently available packages for image processing
The following is a list of currently available packages of image
processing source code. Some packages are public domain, others are
one-time licenses. I would welcome any additions or corrections to
this list. Thank you for all contributions for date.
Scott E. Johnston
johnston@ads.com
Advanced Decision Systems, Inc.
Mountain View, CA 94043
***********
ALV Toolkit
Contact: alv-users-request@uk.ac.bris.cs
Description:
Public domain image processing toolkit written by Phill Everson
(everson@uk.ac.bris.cs). Supports the following:
- image display
- histogram display
- histogram equalization
- thresholding
- image printing
- image inversion
- linear convolution
- 27 programs, mostly data manipulation
***********
BUZZ
Contact: Tehnical: Licensing:
John Gilmore Patricia Altman
(404) 894-3560 (404) 894-3559
Artificial Intelligence Branch
Georgia Tech Research Institute
Georgia Institute of Technology
Atlanta, GA 30332
Description:
BUZZ is a comprehensive image processing system developed at Georgia
Tech. Written in VAX FORTRAN (semi-ported to SUN FORTRAN), BUZZ
includes algorithms for the following:
- image enhancement
- image segmentation
- feature extraction
- classification
***********
LABO IMAGE
Contact: Thierry Pun Alain Jacot-Descombes
+(4122) 87 65 82 +(4122) 87 65 84
pun@cui.unige.ch jacot@cuisun.unige.ch
Computer Science Center
University of Geneva
12 rue du Lac
CH-1207
Geneva, Switzerland
Description:
Interactive window based software for image processing and analysis.
Written in C. Source code available. Unavailable for use in
for-profit endeavours. Supports the following:
- image I/O
- image display
- color table manipulations
- elementary interactive operations:
- region outlining
- statistics
- histogram computation
- elementary operations:
- histogramming
- conversions
- arithmetic
- images and noise generation
- interpolation: rotation/scaling/translation
- preprocessing: background subtraction, filters, etc;
- convolution/correlation with masks, image; padding
- edge extractions
- region segmentation
- transforms: Fourier, Haar, etc.
- binary mathematical morphology, some grey-level morphology
- expert-system for novice users
- macro definitions, save and replay
Support for storage to disk of the following:
- images
- vectors (histograms, luts)
- graphs
- strings
***********
NASA IP Packages
VICAR
ELAS -- Earth Resources Laboratory Applications Software
LAS -- Land Analysis System
Contact: COSMIC (NASA Facility at Georgia Tech)
Computer Center
112 Barrow Hall
University of Georgia
Athens, GA 30601
(404) 542-3265
Description:
VICAR, ELAS, and LAS are all image processing packages available from
COSMIC, a NASA center associated with Georgia Tech. COSMIC makes
reusable code available for a nominal license fee (i.e. $3000 for a 10
year VICAR license).
VICAR is an image processing package written in FORTRAN with the
following capability:
- image generation
- point operations
- algebraic operations
- local operations
- image measurement
- annotation and display
- geometric transformation
- rotation and magnification
- image combination
- map projection
- correlation and convolution
- fourier transforms
- stereometry programs
"ELAS was originally developed to process Landsat satellite data, ELAS
has been modified over the years to handle a broad range of digital
images, and is now finding widespread application in the medical
imaging field ... available for the DEC VAX, the CONCURRENT, and for
the UNIX environment." -- from NASA Tech Briefs, Dec. 89
"... LAS provides a flexible framework for algorithm development and
the processing and analysis of image data. Over 500,000 lines of code
enable image repair, clustering, classification, film processing,
geometric registration, radiometric correction, and manipulation of
image statistics." -- from NASA Tech Briefs, Dec. 89
***********
OBVIUS
Contact: for ftp --> whitechapel.media.mit.edu
otherwise --> heeger@media-lab.media.mit.edu
MIT Media Lab Vision Science Group
(617) 253-0611
Description:
OBVIUS is an object-oriented visual programming language with some
support for imaging operations. It is public domain CLOS/LISP
software. It supports a flexible user interface for working with
images. It provides a library of image processing routines:
- point operations
- image statistics
- convolutions
- fourier transforms
***********
POPI (DIGITAL DARKROOM)
Contact: Rich Burridge
richb@sunaus.sun.oz.AU
-- or --
available for anonymous ftp from ads.com
(pub/VISION-LIST-BACKISSUES/SYSTEMS)
Description:
Popi was originally written by Gerard J. Holzmann - AT&T Bell Labs.
This version is based on the code in his Prentice Hall book, "Beyond
Photography - the digital darkroom," ISBN 0-13-074410-7, which is
copyright (c) 1988 by Bell Telephone Laboratories, Inc.
***********
VIEW (Lawrence Livermore National Laboratory)
Contact: Fran Karmatz
Lawrence Livermore National Laboratory
P.O. Box 5504
Livermore, CA 94550
(415) 422-6578
Description:
Window-based image-processing package with on-line help and user
manual. Multidimensional (2 and 3d) processing operations include:
- image display and enhancement
- pseudocolor
- point and neighborhood operations
- digital filtering
- fft
- simulation operations
- database management
- sequence and macro processing
Written in C and FORTRAN, source code included. Handles multiple
dimensions and data types. Available on Vax, Sun 3, and MacII.
------------------------------
Date: Tue, 20 Feb 90 18:38 EST
From: DONNELLY@AMY.SKIDMORE.EDU
Subject: digital photography
Please help me obtain information about the manipulation of
photographic images digitally.
What are the best products that can be used with a MacIIcx?
Did anyone attend the recent conference on Digital Photography
that took place in Wash DC?
Are there any new interesting products?
Are there any good books on the subject?
Thanks for your assistance.
Denis Donnelly
donnelly@amy.skidmore.edu
------------------------------
Date: Mon, 19 Feb 90 11:33:25 EST
From: flynn@pixel.cps.msu.edu (Patrick J. Flynn)
Subject: CFP: IEEE TPAMI Special Issue on 3D Scene Interpretation
Call for Papers
Special Issue of the IEEE Transactions on
Pattern Analysis and Machine Intelliegnce
on
Interpretation of 3D Scenes
Papers are solicited for a Special Issue of the IEEE Transactions on
Pattern Analysis and Machine Intelligence which will address the
subject of Interpretation of 3D Scenes. The issue is scheduled for
publication in September, 1991. The Guest Editors for the special issue
will be Anil Jain of Michigan State University and Eric Grimson of
M.I.T.
The interpretation of 3D scenes is a difficult yet an important area of
research in computer vision. Advances in sensors that directly sense in
3D and progress in passive 3D sensing methods have resulted in a steady
but not spectacular progress in 3D scene interpretation. The quality of
sensed data is getting better and faster hardware presents more
alternatives for processing it. However, the problems of object
modeling and matching still pose difficulties for general real world
scenes. Problems in 3D sensing, modeling, and interpretation are being
investigated by a number of vision researchers in a variety of
contexts. The goal of the special issue is to gather significant
research results on sensing, modeling, and matching into one volume
which specifically addresses these issues.
Papers describing novel contributions in all aspects of 3D scene interpretation
are invited, with particular emphasis on:
-- 3D sensing technologies, both active (laser, sonar, etc.) and
passive (stereo, motion vision, etc.),
-- 3D object recognition, both from 3D data and from 2D data,
-- 3D navigation and path planning
-- novel object representations that support 3D interpretation
-- applications (e.g. cartography, inspection, assembly, navigation)
-- representation and indexing of large libraries of objects
-- CAD-based 3d vision
-- architectures for 3D interpretation
We particularly encourage papers that address one or more of these
issues or related issues in 3D interpretation, especially in the context
of complex scenes. While both theoretical and experimental contributions
are welcomed, contributions in which new ideas are
tested or verified on real data are specially sought.
All papers will be subjected to the normal PAMI review process. Please
submit four copies of your paper to:
Eric Grimson
M.I.T. AI Laboratory
545 Technology Square
Cambridge, MA 02139
The deadline for submission of manuscript is October 1, 1990. For further
information, contact Anil Jain (517-353-5150, jain@cps.msu.edu)
or Eric Grimson (617-253-5346, welg@ai.mit.edu).
------------------------------
Date: Fri, 16 Feb 90 14:59:58 -0800
From: bertolas@cs.washington.edu (William Bertolas)
Subject: CVGIP TOC, Vol. 50, No. 1, March 1990
Computer Vision, Graphics, and Image Processing
Volume 50, Number 1, March 1990
CONTENTS
M.J. Korsten and Z. Houkes. The Estimation of Geometry and Motion of a
Surface from Image Sequences by Means of Linearization of a Parametric
Model, p. 1.
Clifford A. Shaffer and Hanan Samet. Set Operations for Unaligned Linear
Quadtrees, p. 29.
Phillip A. Veatch and Larry S. Davis. Efficient Algorithms for Obstacle
Detection Using Range Data, p. 50.
David C. Knill and Daniel Kersten. Learning a Near-Optimal Estimator for
Surface Shape from Shading, p. 75.
NOTE
Amelia Fong. Algorithms and Architectures for a Class of Non-Linear
Hybrid Filters, p. 101.
Hug-Tat Tsui, Ming-Hong Chan, Kin-Cheong Chu, and Shao-Hua Kong.
Orientation Estimation of 3D Surface Patches, p. 112.
BOOK REVIEW
Michael Lachance. An Introduction to Splines for Use in Computer
Graphics and Geometric Modeling. By R.H. Bartels, J.C. Beatty,
and B.A. Barsky, p. 125.
ABSTRACTS OF PAPERS ACCEPTED FOR PUBLICATION, p. 000.
------------------------------
Date: 16 Feb 90 20:23 GMT
From: sinha@caen.engin.umich.edu (SARVAJIT S SINHA)
Subject: Conference on Visual Information Assimilation in Man and Machine
Keywords: Conference, Call for Participation, Vision, Information Assimilation
Organization: U of M Engineering, Ann Arbor, Mich.
CALL FOR PARTICIPATION
CONFERENCE ON VISUAL INFORMATION ASSIMILATION
IN MAN AND MACHINE
University of Michigan,
Ann Arbor, MI
June 27-29, 1990
In the last 20 years a variety of computational, psychological and neuro-
biological models of vision have been proposed. Few of these models have
presented integrated solutions; most have restricted themselves to a
single modality such as stereo, shading, motion, texture or color.
We are hosting a 3 day conference be held June 27-29, 1990 at the University
of Michigan, which will bring together leading researchers from each of
these academic areas to shed new light on the problem of how visual
information is assimilated in both man and machine. We have invited
researchers from both academic instituitions and research centers in order
to increase the cross-pollenation of ideas.
Among the questions that we anticipate to be addressed by all
perspectives are: What are the possible stages and representations
for each visual modality? How is contradictory visual information
dealt with? Is there in natural vision systems (and should there be
in computer vision) one coherent representation of the world---a
single model? If a single model will suffice, how (and where in
neurobiology) can visual information be combined into such a model?
If a single model will not suffice, or are there reasons to
believe that there are ways of partitioning visual information
among multiple models that are more likely to be used in man and
useful in machines?
Invited Talks
Irving Biederman (University of Minnesota)
Human Object Recognition
Stephen M. Kosslyn (Harvard University)
Components of High-Level Vision
Whitman Richards (MIT) and Allen Jepson (Univ. of Toronto)
What is Perception?
Geoffrey R. Loftus (Univ. of Washington)
Effects of Various Types of Visual Degradation
on Visual Information Acquisition
Barry J. Richmond (National Inst. of Mental Health)
How Single Neuronal Responses Represent Picture
Features Using Multiplexed Temporal Codes
Patrick Cavanagh (Harvard University)
3D Representation
Daniel Green (University of Michigan)
Control of Visual Sensitivity
Laurence Maloney (New York University)
Visual Calibration
Misha Pavel (Stanford University)
Integration of Motion Information
Brian Wandel (Stanford University)
Estimation of Surface Reflectance and Ambient Illumination
Klaus Schulten (Univ. of Illinois)
A Self-Organized Network for Feature Extraction
John K. Tsotsos (Univ. of Toronto)
Attention and Computational Complexity of
Visual Information Processing
Shimon Ullman (Weizmann Inst-MIT)
Visual Object Recognition
For an extended e-mail announcement, send a message to
iris@caen.engin.umich.edu
For further information contact the University of Michigan Extension Service,
Department of Conferences and Institutes, 200 Hill Street, Ann Arbor, MI
48104-3297. Telephone 313-764-5305.
Sarvajit Sinha sinha@caen.engin.umich.edu
157, ATL Bldg,University of Michigan 313-764-2138
------------------------------
Date: 19 Feb 90 14:08:42 GMT
From: arkin%pravda@gatech.edu (Ronald Arkin)
Subject: VBC - 90 Preliminary Conference Announcement
Keywords: visualization, conference, biomedical
Organization: Georgia Tech AI Group
VBC '90
PRELIMINARY CONFERENCE PROGRAM
Georgia Institute of Technology
and
Emory University School of Medicine
host the
First Conference on
Visualization in Biomedical Computing
May 22-25, 1990
RITZ-CARLTON BUCKHEAD
ATLANTA, GEORGIA
PURPOSE
The goal of the First Conference on Visualization in Biomedical Computing
(VBC) is to help define and promote the emerging science of visualization
by bringing together a multidisciplinary, international group of researchers,
scientists, engineers, and toolmakers engaged in all aspects of scientific
visualization in general, and visualization in biomedical computing in
particular.
THEME
Visualization in scientific and engineering research is a rapidly emerging
discipline aimed at developing approaches and tools to facilitate the inter-
pretation of, and interaction with, large amounts of data, thereby allowing
researchers to "see" and comprehend, in a new and deeper manner, the systems
they are studying. Examples of approaches to scientific visualization include
the dynamic presentation of information in three dimensions, development of
dynamic methods to interact with and manipulate multidimensional data, and
development of models of visual perception that enhance interpretive and
decision-making processes. Examples of visualization tools include graphics
hardware and software to graphically display and animate information, as well
as environments that facilitate human-machine interaction for the interpreta-
tion of complex systems. Examples of applications of visualization in biomed-
ical computing include presentation of anatomy and physiology in 3D, animated
representation of the dynamics of fluid flow, and graphical rendering of bio-
molecular structures and their interactions.
AUDIENCE
The presentations, discussions, and interactions by and between participants
will be of interest to scientists, engineers, medical researchers, clini-
cians, psychologists, and students interested in various aspects of visualiza-
tion.
COOPERATING/CO-SPONSORING ORGANIZATIONS
Alliance for Engineering in Medicine and Biology
American Association of Physicists in Medicine
Emory-Georgia Tech Biomedical Technology Research Center
Emory University School of Medicine
Georgia Institute of Technology
IEEE CS Technical Committee on Computer Graphics
IEEE Computer Society
IEEE Engineering in Medicine and Biology Society
Institute of Electrical and Electronics Engineers (IEEE)
International Federation for Medical and Biological Engineering
International Medical Informatics Association
National Science Foundation
OVERVIEW OF VBC 90
The technical program of VBC 90 will consist of:
o One day of tutorial courses by leading experts
o A plenary session highlighting invited speakers
o Two parallel tracks of contributed papers representing both
theoretical and application areas of visualization in biomedical
computing
o A series of panels on issues of controversy or of current interest,
open for discussions among all attendees
o Technical exhibits by numerous commercial vendors of visualization
technologies
The remainder of the VBC 90 program includes continental
breakfast each morning, refreshment breaks each day, an evening
reception, and dinner accompanied by a laser show at Stone
Mountain. Registrants who wish to do so may also obtain
continuing medical education credit. A tear-off registration
panel is included with this program announcement.
TUTORIALS Tutorial courses take place Tuesday May 22 from 8 AM
through 6:30 PM. Each course lasts one half-day (approximately
four hours) and there are a total of four courses offered from
which each registrant can choose two. The four tutorials are:
Morning Afternoon
Tu1a Volume Rendering Tu2a Biomedical Visualization
Tu1b Human Visual Performance Tu2b Stereoscopic Visualization
Techniques
PLENARY SESSION Invited papers will be presented during the first
morning session (W1) Wednesday at 8:30 AM. The distinguished
speakers and their respective talks are:
Dr. HENRY FUCHS, University of North Carolina
Future High-Speed Systems for Biomedical Visualization
Dr. RICHARD FELDMANN, National Institutes of Health
Visualizing The Very Small: Molecular Graphics
___________________________
TECHNICAL PRESENTATIONS Two parallel tracks of contributed
papers will be offered, representing diverse theoretical and
applications-related research topics in biomedical visualization.
The presentation topics and their respective sessions are
organized as follows:
WEDNESDAY AM
o Volume Visualization (W2a)
o Biomedical Applications I: Cells, Molecules, and Small Systems (W2b)
WEDNESDAY PM
o Models of Visualization (W3a)
o Computer Vision in Visualization I: Segmentation (W3b)
THURSDAY AM
o Artificial Intelligence and Inexact Visualization (T1a)
o Biomedical Applications II: Cardiovascular system (T1b)
o Visual Perception (T2a)
o Biomedical Applications III: Flow and MRI Studies (T2b)
THURSDAY PM
o Human-Machine Interfaces (T3a)
o Systems and Approaches I: System Design (T3b)
FRIDAY AM
o Systems and Approaches II: Algorithms (F1a)
o Computer Vision II: Analysis of Imagery II (F1b)
o Mathematical and Computational Models (F2a)
o Biomedical Applications IV: Treatment Planning (F2b)
FRIDAY PM
o Visualization in Medical Education and General Applications (F3a)
o Biomedical Applications V: Tools and Techniques (F3b)
PANELS Two concurrent panels will take place on the afternoons of
both Wednesday and Thursday. The panels are:
Wednesday Afternoon
o Surface Versus Volume Rendering (W4a)
o Chaos and Fractals in Electroencephalography (W4b)
Thursday Afternoon
o The Role of 3D Visualization in Radiology and Surgery (T4a)
o Visualization in the Neurosciences (T4b)
CONFERENCE REGISTRATION
The registration fee for members of Cooperating/Co-sponsoring Organizations
is $295 prior to March 31. The registration fee after this date is $345. For
non-members, the registration fee is $345 prior to march 31 and $395 after
this date. The special student rate is $50. (Proceedings and reception
tickets are not included at the special student rate, but may be purchased
separately.) The registration fee includes conference registration,
proceedings, reception, refreshments, and other amenities involved in making
this a rewarding learning experience.
TUTORIAL REGISTRATION
The tutorial registration fee is $175 per tutorial for attendees registering
prior to March 31 and $215 for attendees registering after this date.
Attendees will receive the special discounted rate of $275 for two tutorials
before March 31. The special tutorial registration fee for students is $95
per tutorial or $150 for two tutorials prior to March 31, and $125 per
tutorial or $190 for two tutorials after this date. The tutorial registration
fee includes course notes and refreshments.
ACCOMMODATIONS
Hotel arrangements are to be handled by the individual directly with The
Ritz-Carlton Buckhead. To reserve your room, you may call the hotel directly
toll free at (800) 241-3333 or (404) 237-2700. A limited number of rooms
have been made available at the special group rate of $110 single or $119
double (plus tax). Please mention "Visualization in Biomedical Computing."
Reservations should be made as soon as possible but not later than March 31.
DISCOUNT AIR TRANSPORTATION
We have made special arrangements to provide you with a 40% discount off the
normal coach fare, no penalties, on Delta Air Lines. Discounts on restricted
supersaver fares are also available. To make your reservations, call
(800) 288-4446 toll free and refer to "Emory University's Delta File No.
A18445.
IMPORTANT DATES
Early registration: March 15 1990
Special hotel room rate guaranteed through: March 15 1990
------------------------------
End of VISION-LIST
********************
