Recognizing ODIA handwritten scripts
Sanjibani Sudha Pattanayak · 2017 Second International Conference on Electrical, Computer and Communication Technologies (ICECCT) · 2017
This paper deals with Optical character recognition (OCR) for Odia handwritten Characters which deserves more attention from the researchers. In the proposed method, an existing database has been studied. First of all the images are preprocessed. Then the preprocessed images are used for feature extraction. Features are extracted using two simple yet powerful algorithms like shadow feature algorithm and zoning based algorithm. Finally using SVM, the characters are classified which gives accuracy upto 85%.