You know those annoying CAPTCHA things that show up on websites in order to verify that you are a human being and not a spam bot? Well, a bunch of clever people have found an ingenious use for them. See, the CAPTCHA asks you to identify a distorted looking word or string of characters that the character recognition software of a computer can't. Book scanning projects have a problem where computer character recognition software can't always recognise text scanned in from books. By taking this unidentified scanned text and using it for CAPTCHAs, the folk at Carnegie Mellon have found an effective way for people everywhere to help identify the unrecognisable words, thereby facilitating the accurate digital archival of books!
While users of websites still have to go through the hassle of entering a CAPTCHA, at least now it's for a good cause!
No comments:
Post a Comment