A CHINESE-CHARACTER-STROKE-EXTRACTION ALGORITHM-BASED ON CONTOUR INFORMATION

In this paper, we present a new stroke-extraction algorithm that integ rates all levels of contour information including boundary points, dom inant points, corner points, segments, cross-section-sequence graph an d character structure to extract strokes of Chinese characters. In the algorithm, first, the boundary points are extracted, then the dominan t and corner points are detected. Third, the character structure inclu ding singular and regular regions are extracted by the contour informa tion and a modified cross-section-sequence graph (CSSG). Finally, a Be zier curve taking dominant points and corner points as inputs is used to check the continuity of strokes. Experimental results show that the proposed algorithm can correctly extract the strokes up to 95% from p rinted and handwritten test samples based on the human perception. Com pared with a typical thinning approach, the proposed algorithm gives b etter results in terms of both stroke smoothness and the precise numbe r of stroke extractions. (C) 1998 Pattern Recognition Society. Publish ed by Elsevier Science Ltd. All rights reserved.