Google is purchasing a fraud-prevention startup that offers the duel purpose of making the Web more secure while also aiding its Book Search and News Archive projects.
The company is called reCAPTCHA, a startup spun off from a research effort Carnegie Mellon University’s computer science department.
The technology is similar to the classic CAPTCHA
The acronym “CAPTCHA” translates in longhand to: Completely Automated Public Turing test to tell Computers and Human Apart.
CAPTCHA implementations typically ask users to enter a string of distorted letters and numbers displayed at the bottom of Web pages before they can proceed to post a comment, complete a transaction or take some other action.
But Google’s (NASDAQ: GOOG) latest buy has a twist on that classic model. reCAPTCHA harvests many of its challenge words from old books and newspapers. Because the paper and ink have often wrinkled and faded, it can be a challenge for the computers to decipher the letters. The result: reCAPTCHA’s technology helps train computers to become better readers.
The technology is aimed improving a technique known as optical character recognition, which converts digital images into plain text.
“Having the text version of documents is important because plain text can be searched, easily rendered on mobile devices and displayed to visually impaired users,” Google wrote in a company blog post. “So we’ll be applying the technology within Google not only to increase fraud and spam protection for Google products but also to improve our books and newspaper scanning process.”
Google plans to apply the technology to projects like its controversial Book Search initiative to improve its searchable archives. Google has been taking similar steps with newspapers, scanning old issues to make them available online.
Google said that more than 100,000 sites have implemented reCAPTCHA’s technology to fight spam and fraud.