Content-Based Classifying Traditional Chinese Calligraphic Images

Abstract

As traditional Chinese calligraphic (TCC) occupies an important place in the life of modern Chinese, there are a lot of TCC images digitalized and exhibited on the Internet. However, effective classification in them is an imperative problem need to be addressed. The paper proposes a content-based classification scheme that represents the visual content of TCC images by a textural feature set. Four kinds of classifier implemented in the scheme learn the characteristics of fundamental TCC style, art movements and calligraphic artists. The experimental results show that the scheme is capable of classifying the TCC image based on calligraphic artists as well as art movements with an accuracy of greater than 85%.