Via DigitalKoans, a report on new open source OCR software.  Now — someone get busy and train it to read polytonic ancient Greek texts accurately …

This entry was posted in General. Bookmark the permalink.

One Response to OCRopus

  1. I’d be interested in training ocropus to read polytonic Greek, but the main issue that prevents me from doing it is copyright.

    Using copyrighted images of books that are themselves in the public domain to digitally read text from them might be considered as derivative work (as defined by copyright law and so protected by it).

    I’m not a lawyer myself and I’m not really sure that this use would be 100% fair use (not only in the US, but also in the EU).

    But before this question is cleared, I think it makes no sense to spend any time trying to extend ocropus with polytonic Greek (if we have to use the images either from Google Book Search or from the Open Content Alliance).

    Unless we get rights-free images first.

Leave a Reply

Your email address will not be published. Required fields are marked *