Kaspersky Lab patents cutting-edge anti-spam technology

Kaspersky Lab announces the successful patenting of cutting-edge anti-spam technology in Russia. The technology provides efficient, high-level detection of unwanted messages in images.

Spam filters currently have little problem detecting spam text messages. That is why spammers often use stealth technology to hide the text of unwanted messages in images. Filtering graphical spam is far more difficult – before an anti-spam filter can establish whether the text in a message is spam, it must first detect the text in an image. The majority of methods used to detect text in images are based on machine recognition of images. Machine recognition, however, requires uniformity in terms of size, style and the arrangement of symbols. This restriction is exploited by spammers who intentionally distort and create ‘noise’ in images to make detection more difficult.

Kaspersky Lab’s cutting-edge technology was designed to effectively detect text and spam in raster images without the need for machine recognition of images. This approach provides high-speed detection and can recognize text in almost any language. Kaspersky Lab’s new anti-spam technology was developed by Eugene Smirnov. The Federal Service for Intellectual Property, Patents and Trademarks granted the patent on 13 January, 2009.

“On the one hand, the new method is quite good at detecting images that contain text in almost any language,” says Eugene Smirnov, developer of the technology and manager of the Anti-Spam Development Group at Kaspersky Lab. “On the other hand, we don’t attempt to read the text using machine recognition, so the method has sufficiently low resource requirements for it to be used in Kaspersky Lab’s high-performance spam filter.”

The new patented technology is based on a probabilistic and statistical approach. Whether or not an image contains text is determined by the layout of the graphic patterns of words and lines as well as the content of the letters and words in those patterns. Dedicated filters ensure that the system is not affected by noise elements or the fracturing of text within images, while obfuscation techniques used in graphic spam such as warping and rotating are counteracted using a unique method of detecting text lines. The new system can also effectively determine whether detected text is spam by comparing its signature to spam templates contained in databases.

“This invention is very important for the anti-spam industry,” comments Nadezhda Kashenko, Patent Law Group Manager at Kaspersky Lab. “It’s worth pointing out that there are lots of different technologies for detecting spam text messages, but there are very few solutions that can recognise a spam text message in an image. These solutions are very complicated and cumbersome because they have to first find the text in the image and only then decide whether it is spam. Eugene Smirnov’s method is unique. It is a new generation technology, which meant we could assert a patent right for it.”

Kaspersky Lab currently has more than 30 patent applications pending in the US and Russia. These relate to a range of technologies developed by company personnel. Additionally, many of today’s antivirus technologies were developed by Kaspersky Lab and are currently used under license by vendors worldwide, including Microsoft, Bluecoat, Juniper Networks, Clearswift, Borderware, Checkpoint, Sonicwall, Websense, LanDesk, Alt-N, ZyXEL, ASUS and D-Link.