Structural attribute feature code representation and recognition of multifont printed Chinese characters

In this paper, a new structural representation and fuzzy matching scheme ar e proposed for multifont printed Chinese character recognition. A Chinese c haracter is decomposed into eight stroke types. A complete structural attri bute feature codes among different types of strokes are defined and extract ed, which consist of weak and strong primary codes and secondary codes. Wea k and strong primary feature codes depict the global and local spatial rela tionships among different types of strokes respectively, and they are used for a detailed match. h fuzzy matching scheme is used for detailed match be tween an input character and candidate characters. An experiment on 3755 Ch inese characters used daily in multifonts and multisizes shows that our met hod is robust and can achieve high recognition accuracy.