On Thu, May 11, 2006 1:03 pm, Robert Cummings wrote:
Edge detection, noise suppression, and data analysis don't quite equate to recognition. Also 30 years of OCR still requires that the sample be good quality and conform to fairly detectable patterns. If this is so trivial, I await the release of your captcha parser. The spammers would probably pay you millions for it. Where exactly is this bleeding edge, and where can I read more about it? I think you're quite wholeheartedly being naive about the complexity of visual recognition. Prove me wrong.
If you had millions, I would prove you wrong. But, in the meantime I have other stuff to do.
The original poster asked the question -- can it be done. And of course, the answer is Yes.
Visual recognition in its entirety is not what we are talking about here. Instead, we are talking about a very specific and limited problem of how a program can detect known characters in noise -- that -- are also detectable by humans.
Granted, the more complex the image, the more difficult for a program to decipher it, but a CAPTCHA has to be, by definition, detectable by a visually unimpaired human.
Image analysis, enhancement, alteration, and such are better performed by computers than by humans -- that's the reason we developed the software in the first place.
The step between analysis and detection is simply meeting an acceptable error threshold.
However, the reasons why CAPTCHA's still work is that the time required to do the analysis could be better spent elsewhere by spammers.
As for me being naive -- well... either one of us could be -- but that's probably what I get for working in signal analysis (seismic data) since 1975.
As for the bleeding edge, that's obvious, just look to medical imaging. They wish that the detection of their problems were as simple as CAPTCHA.
tedd -- -------------------------------------------------------------------------------- http://sperling.com -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
