Integration of Bangla script recognition support in OCRopus

Abstract

OCRopus is an open source state-of-the-art document analysis and OCR system, featuring pluggable layout analysis, pluggable character recognition, statistical natural language modeling, and multilingual capabilities. The system is being developed with the generous support from Google and other organizations. One of the major goals of OCRopus is to make it multi-lingual. Researchers and developers from different languages involved them with this project to integrate their language or script support. The aim of this thesis is to integrate Bangla script recognition support in OCRopus. The major tasks of this thesis will be: experiment the existing algorithms of OCRopus and check their usability for Bangla script, learn the training and recognition procedures and finally integrate Bangla script related available algorithms within OCRopus to support the complete recognition.