Google Street View algorithm can beat CAPTCHAs
Published 17/04/2014 | 19:28
Google has announced that it is using a new algorithm to find and read house numbers in Street View, and correlate those numbers with existing addresses to pinpoint their exact location on Google Maps.
Unlike earlier algorithms, which separated out image localisation, segmentation and recognition, Google’s new approach integrates these three steps and operates directly on the image pixels.
This system is able to accurately detect and read difficult numbers in Street View with 90 per cent accuracy, according to Google.
However, the algorithm is so effective at deciphering numbers that it can also be used to read CAPTCHA puzzles, which have been used by webmasters for more than a decade to prevent automated software from engaging in abusive activities on their sites.
"Turns out that this new algorithm can also be used to read CAPTCHA puzzles – we found that it can decipher the hardest distorted text puzzles from reCAPTCHA with over 99pc accuracy," said Vinay Shet, product manager for Google reCAPTCHA.
"This shows that the act of typing n the answer to a distorted image should not be the only factor when it comes to determining a human versus a machine."
Google, however, says that it has significantly reduced its dependence on text distortions as the main differentiator between human and machine, and instead performs advanced risk analysis.
This has also allowed it to simplify both its text CAPTCHAs and its audio CAPTCHAs, so that getting through this security measure is easy for humans, but still keeps websites protected, the company said.