A symmetry-based coarse classification method for Chinese characters

Kuo Chin Fan, Wei Hsien Wu, Meng Pang Chung

Research output: Contribution to journalArticlepeer-review


In this paper, we present a novel symmetry-based coarse classification method for the preclassification of printed Chinese characters. The proposed method consists of two main modules, recursive radical extraction, and a symmetry test. The former classifies Chinese characters into ten classes according to the composing structure of the characters. Two classes in the ten classes, left-right, and up-down type characters, contain over 85% of the total characters. The latter performs the symmetry test to determine whether the character, or radical in the ten classes, is symmetric or not. The main purpose of the proposed symmetry-test coarse classification method is to reduce the number of characters in each of the ten classes. Four symmetry features are devised to perform the symmetry test. Experimental results reveal that the proposed method can greatly reduce the number of characters in each class to achieve the coarse classification goal.

Original languageEnglish
Pages (from-to)522-528
Number of pages7
JournalIEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews
Issue number4
StatePublished - Nov 2002


  • Coarse classification
  • Radical extraction
  • Symmetry test
  • Target component

Cite this