This Stanford tool can annoy Captchas

A team of Stanford University researchers has invented a standard way to decode those irksome letters and numbers found in Captchas on many major Web sites, including Visa's Authorize.net, Blizzard, eBay, and Wikipedia, says an article published on News.com.

This is being construed to mean that Captchas, those mingled and difficult-to-read letters and numbers websites often ask users to type in a box to prove they are human, are not as effective security tools as previously imagined.

Says the report: "Their decoding technique borrows concepts from the field of machine vision, which has developed techniques to control robots by removing noise from images and detecting shapes. The Stanford tool, called Decaptcha, uses these algorithms to clean up the image so it can be split into more readily recognized letters and numbers."

According to the article, Decaptcha was able to decode 66 per cent of the Captchas used by Visa's Authorize.net payment site, 70 percent of Blizzard Entertainment's Captchas (whose games include World of Warcraft and Diablo) and 25 per cent of Wikipedia's.

However, many firms that use Captcha technology have defended it, saying that this is just one of the several layers of security tech they deploy and, despite the flaws, Captcha is still useful in preventing things like spam and automated ballot-stuffing in online polls.

sneakers