A Novel Framework for Grading of Writers Using Offline Gurmukhi Characters

Abstract In this paper, we have attempted grading of writers based on offline handwritten Gurmukhi characters written by them. In this work, the features used for classification are based on zoning that has the capability of uniquely grading the characters. Also, samples of offline handwritten Gurmukhi characters from one hundred different writers have been taken in this work. In order to establish the correctness of our approach, we have also considered these characters taken from five Gurmukhi fonts. We have used zoning, diagonal, directional, intersection and open end points, and Zernike moments feature extraction techniques in order to find the feature sets and have used HMM and Bayesian decision making classifiers for obtaining a classification score.