Ok, if the image spams all have a different hash wouldn't that make the Hash function built into Fuzzy OCR useless as well? I'm not sure I buy into that thinking. The hash option in my Fuzzy OCR setup runs pretty well.
Bill ----- Original Message ----- From: Christian Recktenwald To: Bill ; users@spamassassin.apache.org Sent: Wednesday, October 25, 2006 10:06 AM Subject: Re: New RBL idea regarding image spam On Wed, Oct 25, 2006 at 10:00:10AM -0500, Bill wrote: > Couldn't there be an RBL established ... that > maintained the hash of known spam images and forego the wordlist detection? most image spam contains small differences (some flipped pixels, etc) so the hash function will return different results. -- Christian Recktenwald : : citecs GmbH i.L. : [EMAIL PROTECTED] Unternehmensberatung fuer : voice +49 711 601 2090 : Boeblinger Strasse 189 EDV und Telekommunikation : fax +49 711 601 2092 : D-70199 Stuttgart