OCR For Handwritten Marathi Script
Saurabh Murlidhar Tapkir · 2012
Optical Character Recognition which is the original method of character recognition many times gives poor recognition rate due to error in character segmentation. Segmentation is an important task of any OCR system. It separates the image text documents into lines, words and characters. The accuracy of OCR system mainly depends on the segmentation algorithm being used. Segmentation of Handwritten Devanagari text is difficult when compared with Printed Devanagari or Printed English or any other Printed document its structural complexity and increased character set. It contains vowels, consonants. Some of the characters may overlap together. The profile based methods can only segment non-overlapping lines and characters. This paper addresses the segmentation of Handwritten Devanagari text document, the most popular script of Indian sub-continent into lines, words and characters. The proposed algorithm is based on projection profiles. Experimental results it is observed that 100% line segmentation and about 98% character segmentation accuracy can be achieved with overlapping lines, words and characters.