Copy Link
Add to Bookmark
Report
VISION-LIST Digest Volume 11 Issue 31
VISION-LIST Digest Tue Sep 01 11:31:36 PDT 92 Volume 11 : Issue 31
- 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 FTP.ADS.COM
Today's Topics:
People without FTP access can now access the Vision List Archive
Ganapathy's camera location determination algorithm
D-Type skeletonisation ( particle counting)
Request for Range Data with Segmentation
Datacube D52 ASIC: 40MHz Imaging System on a Chip
Real degraded images
Pattern Recognition---textbook list and critical reviews
Fingerprint analysis
Character recognition algorithm needed
Research Associate post (Document image analysis, 3 year duration)
Computer Vision Positions Open
Error in posted CFP announcement!
Workshop on Perceptual Organization and Object Recognition
----------------------------------------------------------------------
Date: 1 Sep 92 02:00:11 GMT
From: fredm@media.mit.edu (Fred G Martin)
Organization: MIT Media Laboratory
Subject: People without FTP access can now access the Vision List Archive
[ Now those of you without ftp access that want to access the anonymous
ftp Vision List Archive at FTP.ADS.COM can do so!
phil... ]
People who don't have access to anonymous FTP can do FTP-by-mail, a
public service operated by DEC. Send a message containing the single
word "help" to "ftpmail@decwrl.dec.com" for instructions.
That's right!! You no longer need FTP access to get at files on
anonymous FTP servers. Using FTP-Mail and the kindness of Digital
Equipment Corp., you send a request script to the DEC server machine,
and it fetches the files for you and mails them to your electronic
mail doorstop! Try it, it works! FTP-Mail will even ASCII-encode and
compress binary files upon request.
Fred Martin | fredm@media.mit.edu | (617) 253-7143 | 20 Ames St. Rm. E15-301
Epistemology and Learning Group, MIT Media Lab | Cambridge, MA 02139 USA
------------------------------
Date: 26 Aug 92 01:36:19 GMT
From: pic@mlb.dmt.csiro.au (Peter Corke)
Organization: CSIRO Division of Manufacturing Technology, Melbourne, Australia
Subject: Ganapathy's camera location determination algorithm
I have some problems with the algorithms published by Ganapathy for
camera location determination. Ganapathy has published several papers
on this problem;
1. Bell Labs technical memo 2/11/84
2. Pattern Recogntion Letters 12/84
3. IEEE Conf. on Robotics and Automation 1984
[2,3] are almost identical, and quite different [1]. My results match
those of Ganapathy, but [1] gives different results to [2] and [3] for
the camera orientation. [2,3] result in rows 2 and 3 being swapped,
and row 2 (after swapping) having the wrong sign. I get this result
for Ganapathy's example, and my own camera calibration data. The
camera location is correct in [1,2,3].
Strat[IU Workshop '84] also uses Ganapathy's example. He doesn't give a
rotation matrix, but obtains the same pan/tilt/swing angles.
There also seems to be an error on the sign of q on p.39 [1].
I'd like to hear from anybody who has investigated these algorithms, and
can perhaps comment on these problems.
Peter Corke, PHONE: +61 3 487-9259
Senior Research Scientist INTERNET: pic@mlb.dmt.csiro.au
CSIRO Div. Manufacturing Technology ACSNET or CSNET:pic@mimir.dmt.oz.au
Locked Bag 9, Preston, Australia. 3072
------------------------------
Date: Fri, 28 Aug 92 09:16:47 GMT
From: vasan@cmch.ernet.in
Subject: D-Type skeletonisation (particle counting)
Hello,
My previous question about application of a D-Type skeletonisation seems
to have been misunderstood as a problem in skeletonisation. Sorry for the
confusion.
I intend to use the D-Type skeletonisation basically for a particle
counting application where each object is reduced to a single pixel
thereby easing the problem in counting.
Reframing my question , here it is
An image such as the following
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 1 1 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
( for that matter, any object with an even numbered width horizontally or
vertically would reduce to the template above)
on the application of the structuring templates ( D-Type )
0 0 x
0 1 1
0 0 x ( and its seven other rotations) would reduce to all zeroes.
How is this problem overcome. One alternative is to let the result of
each application of the element on each pixel affect the next pixel on
in the same iteration. This obviously would not lend itself to parallelisation.
Thanks again for the responses
Ciao
SrinivasanVS ( e-mail : vasan@cmch.ernet.in )
------------------------------
Date: Mon, 31 Aug 92 10:33:26 EDT
From: lerner@division.cs.columbia.edu (Michah Lerner)
Subject: Request for Range Data with Segmentation
I am looking for range data sets that include surface classifications,
for the purpose of comparative analysis in my doctoral dissertation.
The thesis title is "Synergistic Smooth Surface Segmentation". Please
send information to lerner@division.cs.columbia.edu. Thanks.
/Michah
------------------------------
Date: Tue, 25 Aug 92 09:13:00 EDT
From: shep@lenny.datacube.com (Shep Siegel)
Subject: Datacube D52 ASIC: 40MHz Imaging System on a Chip
Datacube D52 ASIC: 40MHz Imaging System on a Chip
Over the past twelve months Datacube has been developing its D52
ASIC for application throughout its product line. The D52 ASIC,
a 225K gate 40MHz video processor, is essentially a 40MHz imaging system
placed on a single die. It includes multiple parallel video paths, transmit
and receive gateways, CPU port, crosspoint switch, look-up table, ALU, and
an extensive image statistics unit.
First silicon is fully functional and new product announcements based on the
multiple parallel 40MHz image pipeline technology are expected shortly.
Aside from the 40MHz pipeline speed, the D52 features a virtual memory map
supporting multiple concurrent video pipelines.
Technical questions: shep@datacube.com
Sales/Marketing: wright@datacube.com
Shep Siegel, Principal Design Engineer
Datacube Inc. 300 Rosewood Dr. Danvers, MA 01923 voice: 508.777.4200
------------------------------
Date: Thu, 27 Aug 1992 12:56:36 UTC+0100
From: Gabriel Cristobal <iodgc33@cc.csic.es>
Subject: Real degraded images
I'm interested to obtain via ftp some examples of images degraded
by different amounts of cloud covers in different regions of the
image for testing purposes. I'm interested in REAL images captured
by a sensor (CCD camera, etc).
Thanks a lot.
Gabriel Cristobal
Instituto de Optica del CSIC
Serrano 121
28006 Madrid (SPAIN)
------------------------------
Date: Fri, 28 Aug 1992 20:36:00 GMT
From: reich@unlinfo.unl.edu (stephen reichenbach)
Organization: University of Nebraska - Lincoln
Subject: Pattern Recognition---textbook list and critical reviews
I am seeking suggestions of a textbook for an advanced pattern
recognition course with an emphasis on image processing and computer
vision. For personal reference, I often still turn to an old standard,
"Pattern Classification and Scene Analysis" by Duda and Hart. It
is still a good book but twenty years old. I would appreciate both
suggestions of books on the subject and critical opinions of the
books. I will be glad to post a summary. Thank you.
Steve Reichenbach (reich@cse.unl.edu)
------------------------------
Date: Sun, 30 Aug 92 13:57:35 EDT
From: anjali@whale.ece.psu.edu
Subject: Fingerprint analysis
Hi,
I m interested in fingerprints recognition and i would really
appreciate if anybody out there could cite any paper references
or info about related work.
thanks in advance.
anjali
email address : anjali@whale.ece.psu.edu
------------------------------
Date: Thu, 27 Aug 1992 16:32:07 GMT
From: Colin Tinto <colint@spider.co.uk>
Organization: Spider Systems Limited, Edinburgh, UK
Subject: Character recognition algorithm needed
I am working on a project at home and need to get hold of an
algorithm that can recognise printed characters from a section of
bitmapped graphics, with reasonable speed.
Can anyone here supply me with any ideas ? Any references to Papers or
Books ? Or even another group that may have more information ?
All help appreciated...
! Colin Tinto ! Time Travel Parcel Deliveries Inc !
! colint@spider.co.uk ! The only courier with previous day service.!
------------------------------
Date: Tue, 25 Aug 92 12:00:35 BST
From: John Oakley <john@spec0.ee.man.ac.uk>
Subject: Research Associate post (Document image analysis, 3 year duration)
Research Associate post (Document image analysis, 3 year duration)
Multimedia Information Systems Laboratory
University of Manchester
We have a vacancy for a creative person to work on a project entitled
"An intelligent system for converting scanned documents to symbolic
form". Relevant research experience is essential and a relevant PhD would be desirable. The post is available now.
The work will invole the design, implementation and evaluation of a
knowledge based system for use in the automatic analysis of scanned
documents. The system will enable an electronic representation of complex
documents, containing a mixture of text, graphics and photographs, to
be automatically derived from paper input. The electronic
representation will conform to an emerging international standard
known as Office Document Architecture (ODA).
Salary will be on the research 1A scale (12,129 to 19,328
pounds, currently under review). To apply, please send me a CV and the name of two referees. E-mail would be fine.
Dr John Oakley
Dept. Electrical Engineering,
University of Manchester,
Oxford Rd,
Manchester M13 9PL.
Tel: 061-275-4534
e-mail: j.oakley@uk.ac.man
------------------------------
Date: Mon, 24 Aug 92 13:31:50 EDT
From: mmlai!whitten@uunet.UU.NET (Gary Whitten)
Subject: Computer Vision Positions Open
Positions in Computer Vision Research
The information technology group of Martin Marietta (corporate
research) Laboratories in Baltimore, Md. is currently seeking
individuals with at least a Master's level (or equivalent) knowledge
of computer vision and image processing. Candidates should have a
Computer Science, EE, Math or Physics backgrounnd and a thorough
knowledge of image feature extraction and analysis techniques. An
understanding of Model Based Vision, including Model Based reasoning
and geometry is preferred.
Candidates should have strong algorithm design and programming
capability and be comfortable using standards based workstations with
Unix, C, and X Windows environments. Additional knowledge of
classification techniques, probability theory, computer graphics, or
neural networks is desirable.
Interested parties should send resumes or address questions to:
Gary Whitten: whitten@mml.mmc.com
or
Monndy Eshera: eshera@mml.mmc.com
Martin Marietta Laboratories
1450 South Rolling Road
Baltimore, MD
21227-3898
TEL.: (410)247-0700 ; FAX: (410) 247-4939
------------------------------
Date: Tue, 25 Aug 92 13:38:25 -0400
From: "Andrew Francis Laine" <laine@ocean.cis.ufl.edu>
Subject: Error in posted CFP announcement!
Dear Postmaster:
I submitted the text below to the vision list and have
been receiving complaints about the following:
1) The name of journal has been omitted from the announcement!
2) The date listed on the Subject line should be 9/15 not 7/15.
Could you please correct these errors as soon as possible.
Thanks,
Andrew Laine
Guest Editor
Journal of Mathematical Imaging and Vision
/********************* Requested Posting *******************/
Dear Colleagues:
An announcement for papers was made at this summer's
NSF/SIAM wavelet conference for a special issue of
the Journal of Mathematical Imaging and Vision (Kluwer),
focusing on wavelet theory and applications.
This message is to inform you that the
call for papers has been extended beyond
the initial deadline.
Sincerely,
Andrew Laine
Guest Editor,
Journal of Mathematical Imaging and Vision
FINAL CALL FOR PAPERS:
EXTENDED DEADLINE: September 15, 1992
Papers focusing on wavelets and their applications including but
not limited to the topics listed below are solicited:
- image pyramids
- frames and overcomplete representations
- multiresolution algorithms
- wavelet-based noise reduction and restoration
- multiscale edge detection
- wavelet texture analysis and segmentation
- wavelet-based fractal analysis
- multiscale random processes
- image representations from wavelet maxima or zero-crossings
- wavelets and neural networks
- wavelet compression, coding and signal representation
- wavelet theory and multirate filterbanks
The aim of the special issue is to emphasize the role of mathematics
as a rigorous basis for imaging science. Regular papers should
present innovative or established mathematical techniques applied
to vision and imaging problems in a novel way, including
any new developments (and problems) in mathematics arising from
these applications. Expository articles, and short
articles (no longer than 10 pages) are also encouraged.
The special issue will be published in early 1993.
Manuscripts should not exceed 40 pages, double spaced.
Please submit three copies of your manuscript to:
Professor Andrew Laine
Department of Computer and Information Sciences
Computer Science and Engineering Building (CSE)
Room 301
University of Florida
Gainesville, FL 32611
Email: LAINE@CIS.UFL.EDU
Phone: (904) 392-1239
Fax: (904) 392-1220
------------------------------
Date: Fri, 28 Aug 1992 09:47:57 -0400
From: fpaas10@blekul11.bitnet
Subject: Workshop on Perceptual Organization and Object Recognition
PERCEPTUAL ORGANIZATION AND OBJECT RECOGNITION
Workshop in Brussels, September 17 and 18, 1992
Perceptual organization and object recognition are two
important achievements of the human visual system. Yet, they are
hard to implement in a computer vision system and they have been
treated quite disjointly in former psychological approaches (e.g.,
Gestalt versus cognitive psychology). Recent work, however,
suggests that perceptual organization may well be the process which
delivers the basic building blocks for object recognition.
The aim of this workshop is to bring together researchers
from a diversity of disciplines who are investigating basic problems
of perceptual organization, which have to be solved by vision
systems as intermediate processes before object recognition can be
achieved. Some of these problems include feature detection and
integration, perceptual grouping, figure-ground segmentation,
contour description, structural description, extraction of invariance,
etc. Our feeling is that considerable progress could be made by
cross-fertilization between these isolated research efforts.
Invited Speakers
Biederman, Boselie, Boucart, Davi, Fahle, Humphreys, Kimchi,
Koenderink, Kolinsky, Lawson, Leeuwenberg, Sekuler, Theeuwes,
Tsal, Van Effelterre, Van Gool, Wagemans, and Watt.
Practical Information
Location: The workshop is held at the Palace of the Royal
Academy of Sciences (Hertogsstraat/Rue Ducale 1, B-1000 Brussels,
Belgium) and starts on Thursday morning (September 17, 9 a.m.)
and lasts till Friday night (September 18, 6 p.m.).
Entrance will be free. Transport and accommodation will not
be arranged by the organizers. Please feel free to pass this
announcement to all who might be interested.
Address: All correspondence should be addressed to Johan
Wagemans, Univ. Leuven, Lab Exp Psy, Tiensestraat 102, B-3000
Leuven, Belgium. tel. 32-16-28.59.69; fax. 32-16-28.60.99; E-mail:
FPAAS10@BLEKUL11.BITNET.
------------------------------
End of VISION-LIST digest 11.31
************************
