Polynomial Representation and Recognition of On-line Tamil Language Characters

Natural Language Characters recognition is a long standing, fundamental problem in pattern recognition. Of the two major issues namely character shape representation and classification, classification issue has been, by and large the subject of most studies. Studies on classification assume the descriptions of shape by basic characteristics such as curvature, tangents, and transform coefficients, are sufficiently expressive to justify focusing on such classification. However, a good representation is as important as a good classifier for achieving higher performance. Within the general subject of character recognition, we have chosen Tamil language characters recognition problem for our study. The Purpose of this research article to is to study the possibility of representing Tamil Language Characters using polynomials and to design a neural network for the recognition. In this research paper we focus particularly on preprocessing and feature extraction of online Tamil Language characters. After analyzing the characters in time domain, we have devised a method of representing the Tamil Language Characters using a polynomial of lower degree. For Tamil Language online characters recognition we have extended a recognition method which has been typically used for recognizing uppercase English characters.