Creation of Devanagari Data Resources and OCR Technology Interchange

Suny At Buffalo, Amherst NY

Investigators

Abstract

This collaborative research project will examine how some of the techniques that have been used for English text recognition can be used with enhanced technologies for OCR of Indian language documents. The proposal will create data resources for testing and evaluating Devanagari recognition systems. Devanagari is the script used by a number of Indian spoken languages, prominent among which are Sanskrit, Hindi and Marathi. It is an alphabetic script, used by over 450 million people around the world. The research will be carried out by teams from the Center of Excellence for Document Analysis and Recognition at SUNY-Buffalo and the Indian Statistical Institute. The project is expected to result in new tools for Devanagari and to stimulate research in other areas of script recognition for NLP, document analysis and information retrieval. --