Results Obtained Using a Simple Character Recognition Procedure on Munson´s Handprinted Data

The number of black points in each of the 25 nonoverlapping square regions of a size-normalized character matrix were used to recognize the 3822 uppercase handprinted alphabetic characters from Munson´s multiauthor data set. The recognition accuracy obtained using a Bayes´ classifier, which assumes statistically independent features, compares favorably with earlier results obtained using recognition systems having complexity comparable to ours. Included are results and a recommendation regarding system evaluation procedures.