Copy Link
Add to Bookmark
Report
VISION-LIST Digest Volume 14 Issue 10
VISION-LIST Digest Mon Mar 13 13:50:39 PDT 95 Volume 14 : Issue 10
- ***** 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 TELEOS.COM
Today's Topics:
FFT C code soon to be available
ANNOUNCE: Arbitrary N FFT C-source, prerelease
Digitizer w/Processor on single board.
Re: Two requests for references
help-contrast,brightness
SGI vision package?
Job Vacancy
Lectureships at University of Surrey
----------------------------------------------------------------------
Date: Sat, 04 Mar 1995 10:53:59 -0100
From: jjn@login.dknet.dk (Jens Joergen Nielsen)
Subject: ANNOUNCE: Arbitrary N FFT C-source, prerelease
Hi
I am going to release the C-source for an arbitrary length FFT routine,
which is highly optimized for speed, within the next few weeks. I would
like some feedback before I do so.
Drop me a line if you are interested, and I will e-mail you an uuencoded
file (50kB), mixfft.zip, containing :
mixfft.c C source for the mix-radix FFT.
fftdemo.c C source for the FFT demo program (PC & UNIX).
demo.exe PC executable of the above.
makefile UNIX make file.
readme.txt General information.
The demo program demonstrates the numerical capabilities and measures the
execution time as well. Compile switches allow code generation in PC and
UNIX environments.
Please note that source code in this packet is copyrighted material,
however non-commercial use of the source code and distribution of the
unaltered mixfft.zip file is free.
What I expect in return:
(1) Bug/problem reports.
(2) Suggestions for improvements.
(3) Experience with other platforms.
(4) Benchmarks (see below).
(5) Other comments (encouragement...) :-)
If you havent got the time to play around with it within the next week
or so, then please wait for the updated release to become available via
anonymous FTP (announced in this newsgroup).
Benchmarks:
The following results were obtained with code generated by the
Microsoft Visual C++ compiler with the following compile options:
/nologo /Gs /G2 /W3 /AH /Ox /D "NDEBUG" /D "_DOS" /FR
and executed on a 50MHz 486DX :
FFT length Time FFT length Time
256 0.013 s 200 0.012 s
512 0.034 s 500 0.026 s
1024 0.059 s 1000 0.054 s
2048 0.154 s 2000 0.148 s
4096 0.274 s 5000 0.343 s
8192 0.692 s 10000 0.698 s
Best regards,
Jens Joergen Nielsen
jjn@login.dknet.dk
------------------------------
Date: Tue, 07 Mar 95 16:15:15 EST
Subject: Digitizer w/Processor on single board.
From: Todd_Jochem@IUS4.IUS.CS.CMU.EDU
Does anyone know of a company that makes a 486 or pentium based board
that has an on-board 24 bit color digitier/display driver? I know of
a couple C40 boards that have this functionality, but not any
486/pentium ones. Thanks,
Todd
------------------------------
Date: Sat, 25 Feb 95 13:25:15 -0500
From: Dave Coombs <coombs@cme.nist.gov>
Organization: National Institute of Standards & Technology
Url: http://isd.cme.nist.gov/staff/coombs/
Tel: 301.975.2865, FAX 301.990.9688, lab 301.975.4080, reception 301.975.3441
Us-Mail: NIST, Intelligent Sys Div, Bldg 220/B124, Gaithersburg MD 20899 USA
Disclaimer: Any opinions are mine; endorsement by NIST is not implied.
Subject: Re: Two requests for references
Ramin, I trust all's well. Camcorder video stabilization algs are
published in the Proc of Consumer Electronics Conf. They tend to be
cheap things that work on thresholded binary images (for cheap HW
implementation) and only handle translation of the image. There are
more sophisticated methods (eg, work by Burt, Anandan, etal;
Chellapa, etal; Black & Jepson; Szeliski; Saund?) and they are
usually at least affine warps and may include "panoramic" mosaicing.
Some end up approaching full motion interpretation, but some can be
done in RT with appropriate HW. Obviously they work best for
scenes lacking motion parallax.
best,
Dave Coombs david.coombs@nist.gov
National Institute of Standards & Technology Tel: (301) 975-2865
Intelligent Systems Division FAX: (301) 990-9688
Building 220 Room B-124 recep: (301) 975-3441
Gaithersburg MD 20899 USA http://isd.cme.nist.gov/staff/coombs/
------------------------------
Date: 13 Mar 1995 21:36:15 GMT
From: avadhana@chervil.ece.ucdavis.edu (Niranjan Avadhanam)
Organization: University of California, Davis
Subject: help-contrast,brightness
hi,
i have a few simple questions regarding the amplitude non-linearity
used at the front end of the human visual system models in image
processing.
i have noticed that the models are implemented as a power law
(b=L^(1/3)) or as a logarithmic law or of the form (b=L/(L+La)).
Is b brightness? if so why is it mentioned as contrast c in the
book by Anil Jain (Digital Image Processing)? i am aware of the
michelson definition of contrast.
if b is brightness, what are the units? to me they look inconsistent
because the power law would give a unit based on luminance (cd/m^2)
whereas the model (L/(L+La)) obviously would give a dimensionless
unit.
it would be of great help to me if some knowledgeable netter would
clarify to me the connection between luminance,brightness,contrast and
the michelson definition of the contrast.
should brightness be the input to the following stages of the visual
system model ie, contrast sensitivity, and the detection process?
assuming that the contrast sensitivity and detection can be combined
as a series of filters, would the output of the filterbank give me an
output in brightness units?
thanks in advance,
niranjan
------------------------------
Date: Tue, 7 Mar 1995 13:06:21 +0100
From: Kristian Simsarian <kristian@sics.se>
Subject: SGI vision package?
Can anyone give me a pointer to a vision package that can
run, without much adaptation, on an SGI Onyx reality2?
Thank you in advance,
Kristian Simsarian Swedish Institute of Computer Science
kristian@sics.se Box 1263, S-164 28 Kista SWEDEN
------------------------------
Date: Wed, 8 Mar 1995 11:36:41 +0100
From: harrison@v4.ee.man.ac.uk (Chris Harrison : 0161-275-5705)
Subject: Job Vacancy
Place: Manchester University, Department of Electrical Engineering
Job: Research Fellow in the Signal Processing group
Time scale: From 1/6/95 for 20 months.
Salary: RA II scale - currently #19,326 to #25,735
JOB DESCRIPTION
Description: We have a vacancy for a person with SIGNIFICANT EXPERIENCE
in hardware engines for image/vision processing such as the Datacube system.
We are currently developing algorithms for a new type of image processing
system and require someone who can:
a) Work with the algorithm designer to optimise the algorithms for
real time hardware implementation.
b) Investigate the hardware available off-the-shelf and specify
a suitable configuration.
c) Commission and test the final system. Some participation in
flight trials will be required.
This work forms part of a significant activity at Manchester University
to develop advanced imaging systems for the aerospace industry. The
successful applicant will be expected to interact with other members of
the group on associated research work.
APPLICATION
To apply please send a full CV including the names of two referees to:
Dr C G Harrison
Electrical Engineering
University of Manchester
Manchester, M13 9PL
Informal enquiries are also welcome and should be sent to
c.g.harrison@man.ac.uk
All enquiries will be treated in the strictest confidence.
------------------------------
Date: Thu, 9 Mar 1995 16:51:24 +0000 (GMT)
Subject: Lectureships at University of Surrey
From: J.Illingworth@ee.surrey.ac.uk
UNIVERSITY OF SURREY
====================
DEPARTMENT OF ELECTRONIC AND ELECTRICAL ENG
===========================================
VISION, SPEECH and SIGNAL PROCESSING RESEARCH GROUP
1 Lectureship in Vision, Speech and Signal Processing: Ref: FE3
1 Lectureship in Image Coding, Compression and Multimedia Systems: Ref: FE4
Reflecting its commitment to excellence in fundamental and applied research
the University of Surrey is launching a major initiative with the creation
of up to 30 Foundation Lectureships to foster development in key academic
areas. Funding is through income from the Universitys Foundation Fund arising
from its successful Research Park. The emphasis of the scheme will be to
provide outstanding scholars with an ideal opportunity to concentrate on
developing their research work with minimal teaching and administrative
duties during the first three years following appointment.
Two of the posts are within the VSSP Group, which is headed by Professor
Josef Kittler. The group has about 40 members and is pursing research
in many areas of image processing, computer vision and pattern recognition.
The group has a large base of computing and many specialised imaging
facilities.
Salaries will be on lecturer scale 14,756 - 25,735 p.a.. Applications in
the form of a curriculum vitae together with the names and addresses of
two referees should be sent to Dr Bernard Weiss, Deputy Head (Research),
Department of Electronic and Electrical Engineering, University of Surrey,
Guildford, GU2 5XH, United Kingdom by 31 March 1995.
email: B.Weiss@ee.surrey.ac.uk.
The University of Surrey is a chartered University with an attractive
campus located at Guildford. It offers a well-funded infrastructure and a
thriving student community. Some thirty miles south-west of central
London, it is within easy reach of both of Londons international airports.
Guildford offers a number of pleasant residential areas, an attractive
regional shopping centre, good schools and a lively cultural environment.
Dr. J. Illingworth, | Phone: (0483) 259835
V.S.S.P. Group, | Fax : (0483) 34139
Dept of Electronic and Electrical Eng, | Email: J.Illingworth@ee.surrey.ac.uk
University of Surrey, |
Guildford, |
Surrey GU2 5XH |
United Kingdom |
For more information plus a picture try:
WWW:http://www.ee.surrey.ac.uk/People/J.Illingworth.html
------------------------------
End of VISION-LIST digest 14.10
************************
