Copy Link
Add to Bookmark
Report
VISION-LIST Digest Volume 13 Issue 30
VISION-LIST Digest Fri Jul 01 10:46:37 PDT 94 Volume 13 : Issue 30
- ***** The Vision List host is TELEOS.COM *****
- Send submissions to Vision-List@TELEOS.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@TELEOS.COM
- Access Vision List Archives via anonymous ftp to FTP.TELEOS.COM
Today's Topics:
Display program for raw images
CALL FOR DATA: First UNIPEN Benchmark of On-line Handwriting Recognizers
New Pattern Recognition Software
NonLinear least Sqares
Seeking C source code for features correspendence
Trajectory matching
Contacts needed in California
Looking for Character Recognition Software
Feature selection
Vision Research Position
CFP: Storage and Retrieval for Image and Video Databases
----------------------------------------------------------------------
Date: 30 Jun 1994 21:14:49 GMT
From: thurn@agni.ece.ucsb.edu (Stefan Thurnhofer)
Subject: display program for raw images
Organization: University of California, Santa Barbara
Keywords: display program raw image
Hi everybody,
I posted this program a few months back in the sci.image.processing
newsgroup. Some people mentioned to me that it might also be of
interest to other groups... So, here we go again:
I have a program here which we are using at UCSB for displaying raw
images (i.e., images that are simply stored as a sequence of bytes or
floats and without header information). This is our preferred format
when developing image processing algorithms. Since some people
from other universities and companies here in southern California
were interested in this program, I decided to make it available
for everybody (I suppose there are more people out there who are
sick of converting there results to raster format and then using
xv or a similar program to display it...)
A brief description of what Xsee is about:
Xsee is an X-windows based display program for images in raw
data format. It can also read in run-length encoded SUN
rasterfiles (1 bit per pixel). Most of xsee's features are
kept very low-level (i.e. the emphasis here is on a program
that can be used for image processing research rather than
artisitic image processing). Basically, all of the available
commands have been added whenever there was a specific use for them
in our lab. This also means that xsee doesn't have those bells
and whistles that some other display programs have.
However, there was a definite need for a program that would
handle raw data.
The key idea behind the concept of xsee is to facilitate the
development of image processing algorithms. To this end, certain features
are quite useful: maintaining the aspect ratio; being able to zoom
in quickly to a very large degree without distorting the actual
pixel values; being able to save and print the image as a whole
or only portions of it.
For any questions or suggestions, please send e-mail to
thurn@amadeus.ece.ucsb.edu.
Stefan Thurnhofer
[ The code is in the Vision List Archive under
SHAREWARE/CODE/Xsee.uuencode . phil... ]
------------------------------
Date: Wed, 29 Jun 94 15:14:18 EDT
From: Stan Janet <stan@magi.ncsl.nist.gov>
Organization: National Institute of Standards and Technology, Gaithersburg, MD
Subject: CALL FOR DATA: First UNIPEN Benchmark of On-line Handwriting Recognizers
- > First UNIPEN Benchmark of On-line Handwriting Recognizers < -
- > Organized by NIST < -
June 1994
* CALL FOR DATA (please post)
At the initiative of Technical Committee 11 of the International Association
for Pattern Recognition (IAPR), the UNIPEN project was started to
stimulate research and development in on-line handwriting recognition
(e. g. for pen computers and pen communicators). UNIPEN provides a platform of
data exchange at the Linguistic Data Consortium (LDC) and is organizing
this year a worldwide benchmark under the control of the US National Institute
of Standards and Technologies (NIST). The benchmark is concerned with writer
independent recognition of sentences, isolated words and isolated characters of
any writing style (handprinted and/or cursive). Although UNIPEN will provide,
in the future, data for various alphabets, this particular benchmark is
limited to letters and symbols from an English computer keyboard . The data
will be donated by the participants.
* Conditions of participation
Participation to the benchmark is open to any individual or institution who
provides a sample of handwriting in the UNIPEN format which contains at least
12,000 characters. The data must be of acceptable quality and donated by
October 1st, 1994 . The database of donated samples will be available for free
to the data donators. Registration material can be obtained by sending email
to Stan Janet at stan@magi.ncsl.nist.gov or via ftp:
ftp ftp.cis.upenn.edu
Name: anonymous
Password: [use your email address]
ftp> cd pub/UNIPEN-pub/documents
ftp> get call-for-data.ps
ftp> quit
* Organizing committee
Isabelle Guyon, AT&T Bell Laboratories, USA
Lambert Schomaker, Nijmegen Institute for Cognition and Information, The Netherlands
Stan Janet, National Institute of Standards and Technologies, USA
Mark Liberman, Linguistic Data Consortium, University of Pennsylvania, USA
Rejean Plamondon, IAPR, TC11, Ecole Polytechnique de Montreal, Canada
------------------------------
Date: Sun, 26 Jun 94 06:20:44 PDT
From: RichardP@eworld.com
Subject: New Pattern Recognition Software
New Pattern Recognition Software for the Macintosh Now Available
Avian Systems, Inc., announces FullPixelSearch, "an interactive pixel
searching system".
Avian Systems, a developer of scientific imaging software, announces the
availability of FullPixelSearch software that allows users to search and find
important regions within TIFF or PICT images on any color-enabled Macintosh
or Power Macintosh.
FullPixelSearch matches patterns at the pixel level, a function which
previously has been available only on mainframe systems. FullPixelSearch
works with a wide range of images and has been used by:
* Scientists to scan satellite photos for mapping agricultural lands
* Physicians to analyze X-Ray, MRI and other biomedical images
* Graphic Artists to touch-up, restore, and enhance digital photographs
* Law enforcement agencies to analyze classified images
FullPixelSearch was developed by Dr. Richard Podolsky, a noted ecologist and
expert on satellite imagery and remote sensing. According to Dr. Podolsky,
head of development at Avian Systems, "FullPixelSearch allows Macintosh
users to perform extensive searching and pattern recognition within a wide
range of imagery. It's functionality presents users with a unique and
powerful set of tools for exploring the details and subtleties in any digital
image."
FullPixelSearch allows the user to simultaneously author a search routine
while viewing an image. The user assembles search templates either by
copying and pasting regions of interest, or pixel by pixel. FullPixelSearch
scans the image and records the location of regions that match the search
template. According to Dr. Dennis Wildfogel, Software Designer at Tandem
Computers, "Search-find-replace commands in a word processor have great
utility and can save enormous amounts of time. FullPixelSearch applies this
same kind of power to digital imagery."
FullPixelSearch features allow users to:
* Locate regions of interest with extensive control of the search criteria.
* Identify regions in any image, and define a search based on those regions.
* Save, as separate files, search templates and the location of all matches
generated.
* Generate histograms for any selected region of an image and display
coordinates and spectral class value for any pixel.
* Specify 100 percent accuracy of the search routine, or relative accuracy at
one percent intervals.
* Generate a "neighbor report" for all pixels of any color enabling the user
to quantify the degree of randomness or uniformity in an image.
Dr. Podolsky is the developer of GAIA (Geographic Access, Image, and
Analysis), an acclaimed remote sensing software package. He is also an
imaging consultant to industry and government.
FULLPIXELSEARCH is available for $895 (early adopters price).
Avian Systems welcomes publisher and distributor inquiries.
For more information on FULLPIXELSEARCH contact:
Avian Systems, Inc., 1275 15th Street Suite 15G, Fort Lee, New Jersey
07024-1929 Tel: (201) 224-2025; FAX: (201) 224-2566, Compuserve =
76460,2615; America On Line = Podolsky, Applelink = PODOLSKY; Internet =
RichardP@eworld.com
------------------------------
Date: Wed, 29 Jun 94 09:12:20 -10752
Subject: NonLinear least Sqares
From: Feisal Mohammed <feisal@uwi.tt>
Hi,
Does anyone know where I can get a constrained nonlinear least squares
solver? I have Andre Tits excellent CFSQP (a constrained Nonlinear
minimax Optimization routine) and I have reformulated my problem to
suit his solver but would like to know if a routine exists that does
constrained nonlinear least squares. I have gone through the netlib
and statlib with little sucess. Any help will be greatly appreciated.
Thanks
- Feisal
Feisal Mohammed (feisal@ldc.uwi.tt)
Mechanical Engineering, University of the West Indies
St. Augustine, Trinidad, W.I. (809)662-2002 x3171
------------------------------
Date: Thu, 30 Jun 94 16:31:21 BST
From: ongep@eee.bham.ac.uk (Mr. Ee Ping Ong)
Subject: Seeking C source code for features correspendence
SEEKING C SOURCE CODE FOR FEATURES CORRESPONDENCE
I had previously posted to Vision-list regarding "C source code for
feature correspondence". However, I have not yet been able to obtain any C
source code for features correspondence. However, I had receive quite a number
of responses interested in obtaining such source codes for their research work.
Therefore, I am reposting this request to vision-list again, hoping that some
people who have such source codes might have missed my posting and will be able
to provide me with such source codes or give me informations on where to get
such souce codes. I would be very grateful if anybody can email me to provide
me with any informations on this.
Very Much Thanks in Advance,
E.P. Ong
School of Electronic & Electrical Engineering
The University of Birmingham
Email : ongep@eee.bham.ac.uk
------------------------------
Date: Mon, 27 Jun 94 13:19:41 +0100
From: Elena Diaz Fernandez <elena@glup.eleinf.uv.es>
Subject:
Dear all:
I am doing research on how vehicles negotiate curves on highways using
image processing to analyze video tapes.
I have already obtained the trajectories of different vehicles. As you
can imagine, when drivers go through a curve, they usually do not
follow the same path but choose different trajectories with different
radius. I would like to classify and recognize the different driver's
behaviour negotiating curves. To do this, I should define a similarity
function that would allow me to identify to what degree two trajectories
are similar.
I think that this function could be dependent on the curvature in each
point or on the area between trajectories (if we represent radius
vs angle).
I would appreciate to hear your opinion about this matter. Do you know
any work or bibliography about shape similarity measures in non closed
planar curves?
Thanhing you in advance, and best regards from Spain!
You may answer preferently to:
e-mail elena@glup.eleinf.uv.es
Elena Diaz
Dpto. Informatica y Electronica.
Universidad de Valencia,
4 Hugo de Moncada,
46010 Valencia, Spain
------------------------------
Date: Fri, 1 Jul 1994 16:46:18 GMT
From: Adrian PRESTON <TE_S343@atlas.king.ac.uk>
Subject: Contacts needed in California
I would be tremendously grateful if someone could supply me with the
e-mail addresses of vision groups, or of vision researchers in or
around the California area which I will be visiting in August.
Specifically, San Diego SU, University College San Diego, USC, and
the University of California, Berkeley.
Thank you very much for your time.
Adrian Preston
Faculty of Technology
Kingston University, UK
email - te_s343@kingston.ac.uk
------------------------------
Date: Wed, 29 Jun 1994 10:48:28 -0400 (EDT)
From: Robert C Boman <rboman@world.std.com>
Subject: Looking for Character Recognition Software
I am looking for character recognition software to use in machine
vision. It can be either commercial or freeware.
Thanks
------------------------------
Date: 28 Jun 1994 00:31:19 GMT
From: rbp@robin.lanl.gov (Rajendra B. Patil)
Organization: Los Alamos National Laboratory
Subject: Feature selection
Hi all:
I am looking for different techniques for feature
selection/construction/ordering for classification and regression in
different areas of
statistics
machine learning
fuzzy logic
Neural networks
genetic algorithms
any pointers appreciated...
raj patil
------------------------------
Date: Thu, 30 Jun 94 15:41:43 EDT
From: TSHIPLEY@vm.temple.edu
Subject: Vision Research Position
RESEARCH POSITION in the Temple Vision Laboratory.
Position is open for either pre or post-doctoral candidate.
Current topics of research in the Vision Lab include: motion
perception, surface and boundary segmentation, Stereoscopic
perception, and Subjective contours. The laboratory is
supported by the National Science Foundation. Position is
ideal for a recent college graduate planning to attend graduate
school in experimental psychology or cognitive science in 1-2
years. However, the position could be changed to postdoctoral
fellowship for a highly qualified applicant. Some computer
programming experience required; prior research experience
desirable.
For more information send e-mail to Thomas F. Shipley
(TSHIPLEY@vm.temple.edu or TSHIPLEY@templevm.bitnet).
------------------------------
Date: Tue, 28 Jun 94 00:02:56 +0000
From: "Wayne Niblack" <niblack@almaden.ibm.com>
Subject: CFP: Storage and Retrieval for Image and Video Databases
IS&T/SPIE Symposium on Electronic Imaging Science and Technology
Feb 5-10, 1995, San Jose Convention Center, San Jose, CA USA
Storage and Retrieval for Image and Video Databases
Wayne Niblack (IBM), Ramesh Jain (UC San Diego)
Large on-line collections of image, video, and multimedia data are
becoming common. These collections have a wide range of applications
in medicine, remote sensing, industry, engineering, digital video
production, education, entertainment, marketing, and on-line information
services. One of the challenges in using this data is how to model,
organize, and index it for efficient storage and retrieval.
This conference will address issues in storage organization and
hierarchies, delivery of real-time data, "intelligent" content-based
retrieval, automatic and semi-automatic methods of image annotation,
definition and computation of features for image and video indexing,
and modelling of image and video data for storage and retrieval.
Papers are solicited in the following areas:
* Image, video, and multimedia databases
* Storage management for large image and video on demand databases
* Intelligent retrieval methods such as content-based
indexing and geometric/structural hashing
* Query by image example and/or sketch
* Indexing by color, texture, shape, layout, and motion
* Visual query languages and user interface for query
* Semantic queries based on image descriptors
* Image and/or digital video applications
* Neural network, feature-based, and information theoretic
approaches to image and video storage, indexing, and retrieval
* Applications and systems of image and video databases in areas
such as biomedical, publishing, journalism, GIS, cataloging,
and education
In addition to the paper presentations, panel discussions on storage
and retrieval of image and video data are planned.
Information for authors:
* Send a 500 work abstract by:
* email - to abstracts@mom.spie.org. (ASCII format),
* fax - to SPIE at 206-647-1445, or
* mail - (4 copies)
IS&T/SPIE EI 95
SPIE, P.O. Box 10
Bellingham, WA 98225 USA
phone: 206-676-3290 telex 46-7053
* Abstracts should include:
* Title
* Authors' Listing (principal author first) - Full names and
affiliations of authors as they will appear in the program
* Authors' address - Complete address, telephone, fax and e-mail
address.
* Conference title: Storage and Retrieval for Image and Video Databases
* Abstract - 500 words
* Biography - for principal author only, 50 - 100 words
* Dates:
* Abstracts due: July 11, 1994
* Acceptance notification: October 28, 1994
* Camera ready abstract due: November 21, 1994
* Manuscript due: January 9, 1995
------------------------------
End of VISION-LIST digest 13.30
************************
