Previous [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]


Journal of Information Science and Engineering, Vol.19 No.4, pp.555-570 (July 2003)

A Coarse Clasification Scheme on Printed Chinese Characters by Encoding the Feature Points

Ming-Gang Wen*+, Chin-Chuan Han++, Kuo-Chin Fan+
and Da-Way Tang+
*Department of Information Management
National Lien-Ho Institutes of Technology
Miaoli, 360 Taiwan
+Department of Computer Science and Informatin Engineering
National Central University
Chungli, 320 Taiwan
++Department of Computer Science and Information Engineering
Chung-Hua University
Hsinchu, 300 Taiwan

In this paper a coarse classification scheme is proposed to speed up the recognition process of machine printed Chinese character. Simple and stable features are extracted by encoding feature points into a codeword of length 16. Geometrically, the codeword represents the distribution of feature points among character strokes. Using these simple features to do coarse classification can eliminate a large number of impossible candidates. Only few surviving candidates need be re-checked in the following more complex recognition algorithms. This scheme will simplify the design of the classifier algorithm and reduce the recognition time. Some experimental results are given to show the validity and efficiency of our proposed methods.

Keywords: coarse classification, feature point extraction, codeword encoding, optical character recognition, threshold-based selection

Full Text () Retrieve PDF document (200307_01.pdf)

Received April 19, 2002; accepted September 25, 2002.
Communicated by Mark H. Y. Liao.