If you've enjoyed decrypting the often frustratingly skewed (and occasionally
humorously juxtaposed) reCAPTCHAs, you might be a bit sad to learn that Google is mixing things up with some rather more boring numerals. The combinations of two words are typically used as part of a registration form to ensure the registrant is, indeed, human. Google is now replacing one of the words in some of its reCAPTCHA forms with photos gleaned from
Street View service. Google says it uses these numbers internally to improve the accuracy of Street View and that pulling them into reCAPTCHAs is part of an "experiment" to "determine if using imagery might also be an effective way to further refine our tools for fighting machine and bot-related abuse online."
In other words, Google's bots are already capable of decoding these numbers, which makes this all sound like a bit of a challenge to the rest of the OCR-loving coders in the world. Any takers?
[Image Credit:
dirtbag]
Google reCAPTCHAs now featuring Street View addresses, 221b Baker St. to get even more famous originally appeared on Engadget on Fri, 30 Mar 2012 08:01:00 EDT. Please see our terms for use of feeds.
Permalink 
TechCrunch |
BlackHat World Forums |
Email this |
Comments
Google first launched its bot-detection reCAPTCHA system in 2007, which means that for over 10 years we've been deciphering garbled text, identifying street lights and clicking tiny boxes in a bid to prove we're human and subsequently access the site...
Google's reCAPTCHA has evolved from distorted text, to street numbers, to "I'm not a robot" tickboxes and, most recently, to their new invisible system. And now the company is bringing its bot-fighting program to Android.
Over the years, Google has utilised a number of methods to distinguish between human and bots on the web. Its take on the CAPTCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) test, known as reCAPTCHA, has required you...
