Character recognition using statistical moments

Abstract This paper presents a character recognition system that is implemented using a variety of statistical moments as features. These moments include Hu moment invariants, Affine moment invariants and Tsirikolias–Mertzios moments. Euclidean distance measure, cross correlation and discrimination cost were used as the classification techniques. The mean of the intraclass standard deviations of the features was used as a weighting factor during the classification process to improve recognition accuracy. The system was rigorously tested under different conditions, including using different number of training sets and documents with different fonts. It was found that Tsirikolias–Mertzios moments with weighted cross correlation classifier provided the best recognition rates.