Copy Link
Add to Bookmark
Report
VISION-LIST Digest 1988 10 19
Vision-List Digest Wed Oct 19 12:34:49 PDT 1988
- Send submissions to Vision-List@ADS.COM
- Send requests for list membership to Vision-List-Request@ADS.COM
Today's Topics:
ever hear of FULCRUM? Help...
Re: Looking for info on Complex Cepstrums
Request for sensor information
----------------------------------------------------------------------
Date: Fri, 14 Oct 88 19:30:08 PDT
From: palmer%hbvb.span@VLSI.JPL.NASA.GOV (Gary Palmer)
Subject: ever hear of FULCRUM? Help...
Hello everyone, I need some help.
I am looking for some (any) information on a
system/package/who-knows-what that does automated mapping. What I
have heard is that one called FULCRUM exists but I have been unable to
gather any other information. That's all I know, any other help would
be greatly appreciated... Any leads about other automated mapping
(from you cartographers out there) products would also be appreciated.
If there are any requests to do so, I will post my findings.
Please respond to me directly as I do not subscribe to these BB's
that I am posting to.
Many Thanks,
Gary Palmer
Science Applications International Corp.
(213) 781-8644
SPAN: hbva::palmer
INTERNET: palmer%hbva.span@vlsi.jpl.nasa.gov
[ I am posting this because many of the problems in automated mapping are
experienced in spatial vision. If you think your answers are relevant
to vision-types, please post to this List.
phil... ]
------------------------------
Date: Mon, 17 Oct 88 12:18:56 EDT
From: efrethei@BLACKBIRD.AFIT.AF.MIL (Erik J. Fretheim)
Subject: Re: Looking for info on Complex Cepstrums
Cepstrums come in two varieties. The first of these is
called the power cepstrum and is defined as the
Fourier transform of the log magnitude of a Fourier
transform.
FFT{log(|FFT{f(x)}|}
This gives a non-invertable result which shows the relative
magnitude of occurrences of frequency intervals in the base
function. A this is used in voice analysis where it will give
a relatively constant spike for gluttal pitches, and at another
point on the quefrency scale give a series of peaks which represent
the formants for any particular utterance. In image analysis the
power cepstrum can be used to discover any regularity of structure.
The complex cepstrum differs from the power cepstrum in that it is
invertable. The complex cepstrum is computed by calculating an FFT
using the magnitude of an FFT as the real term and the phase of an
FFT as the imaginary term. The complication in computing the complex
cepstrum is that the phase must be unwrapped to be continuous, rather
than being limited to values between PI and -PI or 0 and 2*PI. The
complex cepstrum is used in canceling echos in seismographic testing
and for similar functions in voice processing. In image processing it
can be used to cancel multiple copies of an object. For example, if
an image has three vertical bars, one of brightness level 80, one 50, and
one 30, a change to one term in the complex cepstrum will result in an
image with only the 80 bar and slight (2-3 gray levels) disturbances
where the other bars had been. Another tweek to the complex
cepstrum will introduce multiple copies of a bar into a scene where there
previously was only one.
I have been unable to find any references which do more than describe the
complex cepstrum and report it's uses. I am interested in finding
information about the theory behind the complex cepstrum, guidelines on
how to consistently decide just where to apply changes to the complex
cepstrum to obtain consistent results in the image (so far a guessing
game) and information about two-dimensional applications of the complex
cepstrum. I have an image processing problem where if I could get a
consistent theory for the complex cepstrum, I feel I could make some
progress.
------------------------------
Date: Tue, 18 Oct 88 15:42:47 EDT
From: Mark Noworolski <noworol@eecg.toronto.edu>
Subject: Reqest for sensor information
Organization: EECG, University of Toronto
A project that I'm currently on requires that I be able to locate
the area around an individuals open mouth- with his face being well
lit (similar to a dentists light).
My preferred method would be to use some kind of DSP techiniques on
the visual signal received by sensor 'X'.
My current problem is finding said sensor 'X' at a reasonable price.
(nb 2 dimensions are essential- the vision element can not scan left
to right or up and down).
A 2D CCD array would be nice but $ come into play there (or am I mistaken?)
I recall reading in byte about an 'optical ram' sometime around Sept 83.
In fact I even used that part once and it only cost me about $50.
Unfortunately it was very contrast sensitive and quite difficult to get;
and so now I'm considering it as my last resort.
Does anyone know where I can get parts of comparable value for a reasonable
price? Even a general guideline where to look would be appreciated.
Hoping for some help,
mark (noworol@godzilla.eecg or noworol@ecf.toronto.edu)
[ The price of 2d CCD cameras has come down quite a bit, so you might
be able to get one for around $500. If it's 2D CCD imaging chips you're
after (without the camera), you might take a look at Fairchild Weston/
Schlumberger. Looking art their catalog, they have a 488x380 CCD
chip (CCD222) and DSP chips which may solve your problem (phone
number is 408-720-7600 USA). There are many others.
phil... ]
------------------------------
End of VISION-LIST
********************
