A New Method for Character Segmentation and Skew Correction on Chinese Seal Images

In this paper, a new method for automatic character segmentation and skew correction on Chinese seal images is proposed. Traditional methods [1] using shapes of Chinese characters on circular seals sometimes lead to inaccurate transformation of the Chinese characters and thus lose a lot of detailed information. It is hard for these methods to apply Chinese character recognition technology with automatic seal image retrieval. In addition, traditional methods [1] which were designed for circular seal images cannot work well for elliptical seal images. In order to overcome these shortcomings, we propose a new method to segment the Chinese characters on both circular and elliptical seal images. Our method first fits the contour of seal images and then classifies them into either circular seal shape or elliptical seal shape. After that, we apply different methods for Chinese character segmentation and skew correction with circular and elliptical seal shape images respectively. Experimental results show that our method has better performance than traditional methods.