Digitalizing Books Through Captchas
Luis von Ahn Keynotes - In his presentations, Luis von Ahn manages to take difficult subjects such as computer science and...
Captchas are the somewhat distorted words that people have to type out on sites like Ticketmaster.com to authenticate themselves as being a human rather than a program designed to fill out forms and spread spam. Roughly 200 million captchas are being written across the world and if every one of those captchas takes about 10 seconds, that is an incredible amount of time wasted writing out basic words.
Luis von Ahn took this problem and applied to the process of digitalizing books. When a book is to be converted into a digital format, it is scanned and ran through an OCR program which types out the words, some of which it cannot recognize. ReCaptcha is a program that pairs two captchas together, one the computer recognizes and one it doesn't. By typing in both correctly, the computer is able to translate the words it could not recognize by having people translate it for them.