Stroke extraction of Chinese characters plays an important role in Chinese
character information processing such as character recognition, document an
alysis, document compression and storage, font automation and so on. By ana
lyzing the structure of Chinese characters deeply, this paper developed a n
ovel method to extract strokes of Chinese characters directly from the orig
inal character pattern image. Two theorems, eight rules and an algorithm fo
r stroke extraction of Chinese characters are presented. This method can ov
ercome the difficulties encountered in disposing the intersection or connec
tion of different strokes, anti can eliminate noises successfully. Our expe
riments have shown that this method can extract strokes both accurately and
efficiently.