7

From wikipedia

reCAPTCHA is a system ... that uses CAPTCHA to help digitize the text of books while protecting websites from bots

I have a lot of scanned documents that I'd like to convert, and would like to introduce a captcha on my website, so why not kill two birds with one stone?

The reCAPTCHA project has it's own agenda though focusing on archives of The New York Times and books from Google Books.

Does a similar project exist that I could host and thereby dictate the books/docs that are digitised?

Improve this question
11
  • This question is constructive but off topic. It is an exact duplicate of a question on Stack Overflow that predates our site. See this question, stackoverflow.com/questions/244179/…
    – Ben Hoffman
    Commented Oct 7, 2011 at 13:45
  • 1
    @RandomBen I think Craig's looking for a way to use a CAPTCHA-like system to digitise books, not to validate user input (as in that SO question).
    – Nick
    Commented Oct 7, 2011 at 13:50
  • 3
    @Craig My advice would be to use Google's own Optical Character Recognition system to upload your images and convert them to text on the fly (demo and more info). A self-hosted CAPTCHA system would be too slow to convert whole books. reCAPTCHA is only employed to use humans to convert unknown words, not whole documents.
    – Nick
    Commented Oct 7, 2011 at 13:50
  • 2
    @paulmorriss The page you linked to says that the books are first scanned and digitised using OCR, then "reCAPTCHA improves the process of digitizing books by sending words that cannot be read by computers to the Web in the form of CAPTCHAs for humans to decipher." I think the whole point is to translate words OCR can't handle.
    – Nick
    Commented Oct 7, 2011 at 14:15
  • 1
    @Craig - I re-opened the question. This might not be the best site for it but I am not sure which one is at this point.
    – Ben Hoffman
    Commented Oct 7, 2011 at 14:33

1 Answer 1

Reset to default
1

Use Google's OCR to digitize those books. As for using your own books to translate, there isn't currently third-party software available for that. For added reasoning against this, an excerpt from the CAPTCHA Site;

Should I Make My Own CAPTCHA?

In general, making your own CAPTCHA script (e.g., using PHP, Perl or .Net) is a bad idea, as there are many failure modes. We recommend that you use a well-tested implementation such as reCAPTCHA.

Further, he spoke at a TED conference on the subject of reCAPTCHA. If you do infact intend on making your own, might as well study up.

Improve this answer
1
  • The Google OCR link you provided no longer has info on OCR. Do you have an updated link?
    – Bulrush
    Commented May 10, 2016 at 10:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.