The number of black points in each of the 25 nonoverlapping square regions of a sizenormalized character matrix were used to recognize the 3822 uppercase handprinted alphabetic characters from J. H. Munson's multiauthor data set. The recognition accuracy obtained using a Bayes' classifier, which assumes statistically independent features, compares favorably with earlier results obtained. Included are results and a recommendation regarding system evaluation procedures.
ASJC Scopus subject areas
- Theoretical Computer Science
- Hardware and Architecture
- Computational Theory and Mathematics