A method for soliciting help from the general public in order to assist large, text-to-computer projects that digitize thousands of old books. Regular CAPTCHAs are the distorted words found on websites that users type back in to validate that they are humans and not computers. Every day, hundreds of millions of CAPTCHAs are entered, creating a huge pool of human resources to draw from.
In a reCAPTCHA system, the images of words that the optical character recognition (OCR) program cannot decipher are dispersed to several people in the form of a CAPTCHA to get a consensus. Originally developed by Carnegie Mellon University, reCAPTCHA was acquired by Google in 2009. For more information or to get reCAPTCHA code, visit www.recaptcha.net. See OCR
A known word (left) is always sent with the bad word so that the reCAPTCHA serves as a valid CAPTCHA. After several people enter the same text for the bad word, the system considers the word properly converted.
The No CAPTCHA reCAPTCHA
In 2014, Google introduced No CAPTCHA human authentication. Most of the time, users only have to click the box. If the system fails to authenticate, a test is presented such as having to select all the animal pictures from a group that are the same species as the given image.