Automatic Chinese seal identification

A combination of image processing and matching techniques is proposed to solve the problem of seal identification. The techniques include thresholding, rotation, thinning, reexpansion, and finally matching of seal images. The principle for identification is based on the fact that the spatial distributions of the stroke skeletons of the characters engraved by human hands on two distinct seals will never be totally identical in appearance. An appropriate similarity measure for matching stroke skeletons of characters is proposed. A high identification rate of sample seal imprints reveals that the proposed approach is effective.