作者: Y. X. Gu , Q. R. Wang , C. Y. Suen
DOI: 10.1109/TPAMI.1983.4767349
关键词: Tree (data structure) 、 Pattern recognition 、 Computer science 、 Chinese characters 、 Decision tree 、 Orthogonal coordinates 、 Classifier (UML) 、 Artificial intelligence 、 Tree structure 、 Computational Theory and Mathematics 、 Software 、 Applied mathematics 、 Computer Vision and Pattern Recognition
摘要: A multistage classifier with general tree structure has been developed to recognize a large number of Chinese characters. simple and efficient method classifying the characters was achieved by choosing best feature at each stage tree. The features used are Walsh coefficients obtained from two profiles character projected onto X-Y orthogonal axes. Some algorithms for aligning were compared one them adopted in this recognition scheme. high rate about 99.5 percent an experiment more than 3000 different