Development of OCR Techniques for Handwritten Bangla Text

Optical Character Recognition (OCR) involves machine recognition of characters from digitized images of optically scanned document pages. OCR systems facilitate large scale document transcriptions and have potential applications in extracting data from filled-in forms,interpreting handwritten addresses from mail pieces, and so on. An OCR technique is either aimed at printed or handwritten characters of a particular script. The present work addresses certain key problems related to OCR of handwritten documents of Bangla script with successful development of techniques for skew angle detection and text line extraction, word segmentation, and digit, character and word recognition. This work also includes development of applications for recognition of PIN codes from Indian Postal documents, interpretation of handwritten form documents, and an user interface for editing machine recognized handwritten Bangla documents. Bangla is the second most popular script and language in the Indian subcontinent. As a script, it is used for Bangla, Ahamia and Manipuri languages. Bangla, which is also the national language of Bangladesh, is the fifth/sixth most popular language in the world.