Copy Link
Add to Bookmark
Report
VISION-LIST Digest Volume 10 Issue 09
VISION-LIST Digest Fri Feb 22 09:42:47 PDT 91 Volume 10 : Issue 9
- Send submissions to Vision-List@ADS.COM
- Send requests for list membership to Vision-List-Request@ADS.COM
- Access Vision List Archives via anonymous ftp to ADS.COM
Today's Topics:
Motion & Active Vision
Look for email addresses of Haddon and Boyce
Remote Sensing Software
Needed -> Information on printing nice optical flow vectors
NATO Advanced Research Workshop: SHAPE in PICTURE
Paper available -- visual orientation multiplexing
----------------------------------------------------------------------
Date: Mon, 18 Feb 91 08:05:58 GMT
From: eba@computing-maths.cardiff.ac.uk (Veisel Aslantas)
Subject: Motion & Active Vision
Organization: Univ. of Wales Coll. of Cardiff, Dept. of Electronic & Systems
Engineering
Dear friends!
I am a Phd research student involved in Motion and Active Vision( active
perception or dynamic vision, focusing, tracking and eyes convergence, etc.)
Is there anybody who could send to me bibliography references and comments in
this area?. Many thanks in advance, Veisel.
------------------------------
Date: Tue, 19 Feb 91 10:25:30 -0600
From: bho@aio.jsc.nasa.gov (Bill Ho)
Subject: look for email addresses of Haddon and Boyce
Hi,
I am looking for email addresses and telephone numbers of
John F. Haddon and James F. Boyce.
They published a paper on IEEE Transactions on Pattern Analysis
and Machine Intelligence, Vol. 12, No. 10, October 1990.
The title of this paper is "Image Segmentation by Unifying
Region and Boundary Information". I am interested in this paper
and want to ask them some questions.
John F. Haddon is with the Royal Aerospace Establishment,
Farnborough, Hampshire GU14 6TD, England.
James F. Boyce is with Wheatstone Laboratory, King's College,
Strand, London WC2R 2LS, England.
If you know them, please let me know.
I am working for Automation and Robotics Division of NASA/Johnson
Space Center in Houston, U. S. A.
Thanks,
Bill Ho/ER2/JSC/NASA
713-483-4936
bho@aio.jsc.nasa.gov
------------------------------
Date: Tue, 12 Feb 91 10:43:05 +0100
From: ebs@fmi.kth.se (Eberhard Guelch)
Subject: Remote Sensing Software
Hello!
Does anyone have any information on a public domain program-package
for remote sensing tasks like image classification? It must be able
to run on PC-compatibles. I do not need the display part, but only the
image processing part. The program will be used in education in a
third world country, which cannot afford any fancy things. It would
be nice if it can do things like maximum likelihood classification or
similar. C code under Microsoft is preferred. All information is
highly appreciated. Send replys to: pax@fmi.kth.se
------------------------------
Date: Thu, 21 Feb 91 03:54:21 GMT
From: martin@eola.cs.ucf.edu (Glenn Martin)
Subject: Needed -> Information on printing nice optical flow vectors
Hi. I have implemented several optical flow programs, but was wondering how
people prepare the flow diagrams so that they look really nice. For instance,
is postscript usually used, or is there something else? I have no trouble
generating the vectors themselves in a flow diagram but when I use our
current print routine, they come out poorly. An explanation of how you
do it would be sufficient. Thanks.
Glenn Martin
University of Central Florida
martin@eola.cs.ucf.edu
------------------------------
Date: Tue, 19 Feb 91 13:22:24 +0100
From: toet@izf.tno.nl (Lex Toet)
Subject: NATO Advanced Research Workshop: SHAPE in PICTURE
SHAPE in PICTURE
NATO Advanced Research Workshop
A. Toet, Institute for Perception TNO, Kampweg 5,
3769 DE Soesterberg, The Netherlands
FAX: 31-3463-53977 e-mail: toet@izf.tno.nl
H.J.A.M. Heijmans and O Ying-Lie , Centre for Mathematics and Computer Science,
Kruislaan 413, 1098 SJ Amsterdam, The Netherlands
FAX: 31-20-5924199 e-mail: henkh@cwi.nl
The workshop is expected to be held in late 1992, in a recommended
location in southern Europe. The main objective is to gather a
limited number of scientists who are actively involved or very much
interested in the development of mathematical description of shape in
greylevel images.
The duration of the workshop will be about 5 days. There will be
morning sessions for presentation of papers, and afternoon sessions
for posters and further discussions. On demand, evening forums
devoted to selected topics will be organised to allow exhaustive
discussions. Adequate time for breaks will be scheduled between
sessions. The emphasis of the workshop is the exchange of research
and thoughts in a pleasant and stimulating atmosphere.
Participants are expected to make an active contribution by presenting
papers, posters, and/or by joining discussions. Papers may be
presented as a long or short paper. Posters must be accompanied by a
short presentation. Contributions on work in progress are
specifically encouraged.
Proceedings of the workshop will be published according to NATO ARW
regulations. The proceedings shall include the papers, posters,
(edited) discussions, and perhaps conclusions resulting from the
workshop.
***************
Various fields of image analysis require a description of shape. The
images under consideration are two-dimensional distributions of
greylevel values. Recognition and classification of particular shapes
in images is frequently performed by human perception. Most
conventional algorithms suffer from severe shortcomings. The method
may be improved by a suitable description of shape.
The description of shape should reflect the relevant features with
regard to the aimed purpose. In a sense, shape may be formulated as
structured features. These features may be described by a symbolic
representation. Features can be transformed to symbols through a
suitable mapping.
In general, the shapes under consideration have a limited extent.
Thus, the transform should be local, that is, its domain of influence
is restricted. A hierarchical representation of the image may be
obtained by defining an order relation on the describing symbols. Such
a representation allows the analysis of the image at different levels
of "resolution".
A suitable description may satisfy the following properties:
- based on an underlying topology,
- hierarchical in the sense that there is an order relation on
the describing symbols,
- local, that is the domain of influence must be restricted,
- (semi) continuous with respect to the topology,
- invariant under certain affine transforms, such as
translation or rotation,
- compatible with change in scale of both the domain and the
greylevel.
A number of methods of interest are mathematical morphology,
wavelet transform, fractal geometry.
Recently, a formal theory of shape recognition has been introduced.
Categorial shape theory is currently still under development. Formal
theories in computing science might be useful in the definition of the
symbolic representation of shape, and the final implementation on a
computer system.
The construction of the description should involve the use of the
above mentioned methods or other advanced mathematical methods that
are theoretically well funded. It may be posed, that the subject is a
meeting point of mathematics, computing science, and application
fields of image analysis.
***************
Please indicate whether you are interested in participation and
the degree of your commitment as indicated below
O co-editor
O reviewer
O chairperson
O long paper
O short paper
O poster
O tentative
O reasonably sure
O definite
Name :
Affiliation :
Address :
Country :
FAX :
e-mail :
In case you know someone who might be interested, please let us
know.
Please e-mail or fax this part of the form to:
A. Toet, Institute for Perception TNO, Kampweg 5,
3769 DE Soesterberg, The Netherlands
FAX: 31-3463-53977 e-mail: toet@izf.tno.nl
------------------------------
Date: Fri, 22 Feb 91 11:04:31 -0500
From: Jonathan Marshall <marshall@cs.unc.edu>
Subject: Paper available -- visual orientation multiplexing
**** Please do not re-post to other bboards. ****
Papers available, hardcopy only.
ADAPTIVE NEURAL METHODS FOR MULTIPLEXING ORIENTED EDGES
Jonathan A. Marshall
Department of Computer Science
University of North Carolina at Chapel Hill
Edge linearization operators are often used in computer vision and in
neural network models of vision to reconstruct noisy or incomplete
edges. Such operators gather evidence for the presence of an edge at
various orientations across all image locations and then choose the
orientation that best fits the data at each point. One disadvantage
of such methods is that they often function in a winner-take-all
fashion: the presence of only a single orientation can be represented
at any point; multiple edges cannot be represented where they
intersect. For example, the neural Boundary Contour System of
Grossberg and Mingolla implements a form of winner-take-all
competition between orthogonal orientations at each spatial location,
to promote sharpening of noisy, uncertain image data. But that
competition may produce rivalry, oscillation, instability, or mutual
suppression when intersecting edges (e.g., a cross) are present. This
"cross problem" exists for all techniques, including Markov Random
Fields, where a representation of a chosen favored orientation
suppresses representations of alternate orientations.
A new adaptive technique, using both an inhibitory learning rule and
an excitatory learning rule, weakens inhibition between neurons
representing poorly correlated orientations. It may reasonably be
assumed that neurons coding dissimilar orientations are less likely to
be coactivated than neurons coding similar orientations. Multiplexing
by superposition is ordinarily generated: combinations of intersecting
edges become represented by simultaneous activation of multiple
neurons, each of which represents a single supported oriented edge.
Unsupported or weakly supported orientations are suppressed. The
cross problem is thereby solved.
[to appear in Proceedings of the SPIE Conference on Advances in
Intelligent Systems, Boston, November 1990.]
Also available:
J.A. Marshall, "A Self-Organizing Scale-Sensitive Neural Network."
In Proceedings of the International Joint Conference on
Neural Networks, San Diego, June 1990, Vol.III., pp.649-654.
J.A. Marshall, "Self-Organizing Neural Networks for Perception of
Visual Motion." Neural Networks, 3, pp.45-74 (1990).
= Jonathan A. Marshall marshall@cs.unc.edu =
= Department of Computer Science =
= CB 3175, Sitterson Hall =
= University of North Carolina Office 919-962-1887 =
= Chapel Hill, NC 27599-3175, U.S.A. Fax 919-962-1799 =
------------------------------
End of VISION-LIST digest 10.9
************************
