At 8:20 PM -0400 3/29/07, TG wrote:
> Not as hard as you might think. You don't have to identify it as a
pig but rather as the spectral properties that a pig image displays.
> It's like part recognition on an assembly line.
Are you incinerating pigs and doing spectral analysis on them to see what
they're composed of, again?
No, it's a technique that I read about where an assembly line may
have more than one type of object and the object can be in any
orientation. The problem is that they need to identify and count each
object that passes by automatically.
Now, taking an image of the object in question and then comparing
that image to every possible orientation would be very time
consuming. As such, they developed another method.
The method goes like this -- take picture of the object, run it
through an FFT and get it's spectral properties (frequency envelope
-- amp v frequency). Then compare that envelope with envelopes of
unknown objects to see if they match. You see, with the correct
lighting and such, the spectral properties of an object will be the
same regardless of it's orientation. Neat huh?
I did something like that many years ago using EKG's. It provided an
intelligent system that would flag possible heart disease by
comparing the unknown heart EKG (the patient) to known EKG anomalies
(known heart diseases). Rather leading edge at the time -- but, old
stuff now. However, there's still a lot to do if anyone wants to hire
me for this type of research. I hate it when employers get something
worthwhile, think that's all there is, and then vanish without
investigating it further. There's so much more. Oh well, they have
the money and I don't.
Cheers,
tedd
--
-------
http://sperling.com http://ancientstones.com http://earthstones.com
--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
