ONLINE HANDWRITTEN CHINESE CHARACTER-RECOGNITION DIRECTED BY COMPONENTS WITH DYNAMIC TEMPLATES

Structural matching has been recognized as a promising approach for on -line Chinese character recognition. In order to reduce its great comp utational complexity and improve its performance, people have been see king for ways to direct the matching of a whole character by the resul t of partial matching. In this paper, the authors proposed 45 basic co mponents for 3,755 categories of daily-used Chinese characters to dire ct the stroke segment matching of whole characters. Since they are alw ays located at either the beginning or the end of the stroke segment s tring of characters, these components are easy to be extracted and sep arated from other parts of a character. Besides, in our approach, the reference templates of these components are extracted dynamically from the corresponding segment string of characters when a specific matchi ng is carried out. This strategy avoids building multiple templates fo r the components of the same kind but at different places of character s. The experiments show that the segment matching computation has been reduced greatly without reducing the correctness of matching.