A novel neural net-based system and process for constructing optimized prototypes for character recognition

摘要

A system and process for performing character recognition is disclosed wherein inputted characters are compared to prototypes maintained in a predetermined database of the system to determine the best matching character. The prototypes are optimized to improve the recognition capability of the database. Experiments on a large set of handwritten Chinese characters with in average 200 samples per character have shown the present method outperforms the famous LVQ2 algorithm.