Hacker News new | ask | show | jobs
by bob_theslob646 3104 days ago
Please correct me if I am wrong, but this looks like you have to "name" each page. I would also want to see how accurate the ocr is. Historically, ocr on handwritting has been a problem unless the data is perfectly formatted. I guess the case is just to get enough accuracy so that you can look for or at the image of that page with the indexed search term you were looking for.