Affiliation:
1. Shanghai Jiao Tong University
Abstract
The feature points extraction plays an essential role in modeling structure of Chinese characters for accurate recognition. The distortions caused by the thinning process make the thinning-based stroke segmentation difficult for extracting feature points and handle the ambiguous intersection regions. This paper proposes a novel method that overcomes the distortions to extract feature points from the thinned Chinese characters. Firstly, we use eight structural elements to thin a Chinese character in a series combined way, and then segment different kinds of strokes according to their properties after thinning preprocessing. Finally we draw the end points of every stroke based on the Blob algorithm and obtain the cross information at the same time. The feature points will be abstracted fast and exactly. The experimental results show that the proposed method is effective for Chinese characters recognition (CCR).
Publisher
Trans Tech Publications, Ltd.
Reference9 articles.
1. L. Bin and Z. Shuxiang, A New approach to recognition of both handwritten and multi-font printed Chinese characters, Proc. 8th Int. Joint Conf. Pattern Recognition, pp.641-643, (1986).
2. K. W. Gan and K. T. Lua, A new approach to stroke and feature point extraction in Chinese character recognition, Pattern Recognition Lett., vol. 12, pp.381-387, (1991).
3. C. W. Liao and J. S. Huang, Stroke segmentation by Bernstein-Bezier curve fitting, Pattern Recognition, vol. 23, No. 7, pp.475-484, (1990).
4. X. M. Sun and M. J. Yang, A totally knowledge-based stroke extraction method for Chinese characters, Journal of computer research and development. vol. 37, No. 5, pp.543-550, (2000).
5. C. T. Chuang and Y. T. Liu, A Stroke Extraction Method for Multifont Chinese Characters Based on the Reduced Special Interval Graph., IEEE TRANSACTIONS, vol. 25, No. 7, pp.1171-1178, (1995).