reCAPTCHA assisted OCR for Devanagiri Texts
Kailash Atal, Ashish Arora, Devendra Singh Sachan, Prabin Kumar Bora · 2013
CAPTCHA is a challenge response implemented using distorted characters on web to determine whether a user is a human or a computer. reCAPTCHA is constructive use of this human effort to digitize text from old documents which is difficult for OCR and also authenticate human user. The other paradigm is an integrated OCRreCAPTCHA system where OCR digitizes documents with high accuracy and comes up with a confidence score so that characters with low score are digitized using human response. This work is first attempt to build reCAPTCHA assisted OCR system for Devanagiri script that learns from human response on the web and digitizes document with high accuracy.