Decomposition of Chinese character into strokes using mathematical morphology

In off-line character recognition, reliable extraction of strokes greatly i nfluences the overall performance of the system. We propose a novel method of decomposing a Chinese character into a set of strokes, based on the conc epts of mathematical morphology. A character is segmented after being thinn ed using morphological operation. Each segment is expanded by morphological operations such as elongation, fattening and isotropic expansion then inte rsected with the original input shape. The expanded segments are merged usi ng a convexity measure at the final stage. The final merged segments are re garded as basic strokes of a character. To show the effectiveness of the me thod, we perform experiments and the results show the validity of the propo sed method. (C) 1999 Elsevier Science B.V. All rights reserved.