OSDT: Outer Shape Detection Technique for Recognition of Bangla Optical Character

Abstract

Optical Character Recognition is one of the challenging fields in recognition of printed Bangla text. The main difficulties are that there are no precise techniques or algorithms for separating lines, words, and characters from printed Bangla text and efficiently recognize these separate characters. In this paper, we introduce new methods to separate lines, words, and characters from printed Bangla text. We also propose Outer Shape Detection Technique (OSDT), a new technique to recognize each character based on its outer shape which is unique. To successfully accomplish this, proposed technique scans each character both from its left side and right side. Finally our experimental results are compared with some prevalent ones which precisely show that our approach performs smoothly on different font sizes and number of characters.