Character recognition of the Batak Toba alphabet using signatures and simplified chain code

Abstract

This paper presents the simplified chain code (SCC) from our concept of basic shape to recognize the characters of Batak Toba alphabet. The proposed chain code is based on the eight-direction Freeman chain code from which the horizontal and vertical directions are omitted. Hence we focus on shapes with dominant diagonal contours which characterize the recognized alphabet. Our implementation shows that recognizing the alphabet using the scale-invariant SCC can achieve more than 90% accuracy and is more than 10 times faster than that using signatures. To the best of our knowledge, this is the first reported attempt of recognizing the Batak Toba alphabet.