COMPUTER RECOGNITION OF PRINTED BANGLA SCRIPT

This paper considers optical character recognition (OCR) of Bangla, th e second most popular script in the Indian subcontinent. A complete OC R system is described for documents of single Bangla font, where more than three hundred character shapes are recognized by a combination of template and feature-matching approach. Here the document image captu red by a flatbed scanner is subject to tilt correction, line, word and character segmentation, simple and compound character separation, fea ture extraction and finally character recognition. Some character occu rrence statistics have been computed to aid the recognition process. T he simple character recognition is done by a feature-based tree classi fier, and the compound character recognition involves a template match ing approach preceded by a feature-based grouping. At present, recogni tion accuracy of about 96% is obtained by the system.