Role of Skeletonization and Segmentation for Recognition of Off Line Odia Characters

Abstract

Automatic character recognition is becoming an extremely important concept in digital preservation of cultural heritage. The importance of OCR for Indian languages cannot be underestimated as massive volumes of print-media resources representing the cultural and historical heritage of India are available in several Indian scripts. Several recognition techniques for printed characters of different Indian scripts have been reported in literature like Devanagari Tamil, Telugu, Bangla scripts etc. A substantial amount of work has been done in the field of OCR for the above cited scripts but very little research has been done for the Odia language. Moreover whatever the techniques that have been implemented in the past have not yet provided a reliable output. With many different types of algorithms present for different languages, it gave us an opportunity to study various pre-processing, feature extraction and image classification techniques. The basic emphasis is given mainly upon the skeletonization and segmentation to identify several structural behavior of printed odia scripts. newline