Submitted on May 12, 2008
Accepted on August 5, 2008
reCAPTCHA: Human-Based Character Recognition via Web Security Measures
Luis von Ahn 1*, Benjamin Maurer 1, Colin McMillen 1, David Abraham 1, Manuel Blum 1
1 Computer Science Department, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA.
* To whom correspondence should be addressed.
Luis von Ahn , E-mail: biglou{at}cs.cmu.edu
CAPTCHAs (Completely Automated Public Turing test to tell Computers and Humans Apart) are widespread security measures in the World Wide Web that prevent automated programs from abusing online services. They do so by asking humans to perform a task that computers cannot yet perform, such as deciphering distorted characters. Our research explored whether such human effort can be channeled into a useful purpose: helping to digitize old printed material by asking users to decipher scanned words from books that computerized optical character recognition (OCR) failed to recognize. We showed that this method can transcribe text with word accuracy over 99%, matching the guarantee of professional human transcribers. Our apparatus is deployed in over 40,000 Web sites and has transcribed over 440 million words.
