http://www.kiwiblog.co.nz/2006/11/dpf_on_spam_on_radio_nz.html DPF's Kiwiblog - Fomenting Happy Mischief since 2003 Mon, 13 Feb 2012 19:45:56 +0000 hourly 1 http://wordpress.org/?v=3.3.1 http://www.kiwiblog.co.nz/2006/11/dpf_on_spam_on_radio_nz.html#comment-251507 Falafulu Fisi Mon, 06 Nov 2006 09:41:26 +0000 http://kiwiblog-test.inspire.net.nz/wordpress/?p=15190#comment-251507 Legislation will not kill spamming emails. Better technology in anti-spamming will kill off spamming. It is a waste of time devising legislation to thwart spammers. Now the tactics of spammers have now shifted to embedding text on an image which text-spamming filters are useless to stop. The whole field of intelligence image recognition (computer vision) have advanced within the last 5 years or so, that the technology & mathematics to recognise and extract embedded texts in a GIG image is already available. The extracted text then can run thru the normal bayesian text filter. Also the technology to recognize pornographic images are now available. These technologies had been published and appeared in different computing & engineering peer review journals. I am a bit surprised that a security expert from Symantec was interviewed in the 'Close Up' program on TV1 tonight and his answer to the image spamming is to block off any image attachment from emails whether it is legitimate one or a spam mail. Symantec is a huge corporation with unlimited resources and a huge R&D team that I am sure that one of their researchers member might have read or scour papers from journals in 'Pattern Recognition', 'Computer Vision' , 'Image Processing' to find the answer to this problem because there have been publications already available of how to tackle the problem. Legislation will not kill spamming emails. Better technology in anti-spamming will kill off spamming. It is a waste of time devising legislation to thwart spammers. Now the tactics of spammers have now shifted to embedding text on an image which text-spamming filters are useless to stop. The whole field of intelligence image recognition (computer vision) have advanced within the last 5 years or so, that the technology & mathematics to recognise and extract embedded texts in a GIG image is already available. The extracted text then can run thru the normal bayesian text filter. Also the technology to recognize pornographic images are now available. These technologies had been published and appeared in different computing & engineering peer review journals. I am a bit surprised that a security expert from Symantec was interviewed in the ‘Close Up’ program on TV1 tonight and his answer to the image spamming is to block off any image attachment from emails whether it is legitimate one or a spam mail. Symantec is a huge corporation with unlimited resources and a huge R&D team that I am sure that one of their researchers member might have read or scour papers from journals in ‘Pattern Recognition’, ‘Computer Vision’ , ‘Image Processing’ to find the answer to this problem because there have been publications already available of how to tackle the problem.

]]> http://www.kiwiblog.co.nz/2006/11/dpf_on_spam_on_radio_nz.html#comment-251506 GPT Sun, 05 Nov 2006 12:40:37 +0000 http://kiwiblog-test.inspire.net.nz/wordpress/?p=15190#comment-251506 Goodness is nothing off the record these days? :) Goodness is nothing off the record these days?

]]>