A new nonlinear shape normalization method for off-line handwritten Chinese character recognition
How to correct the shape variations of handwritten Chinese characters is very important in off-line handwritten Chinese character recognition. In order to get rid of shape variations and reduce within-category variances directly from handwritten Chinese character image, a new nonlinear shape normalization method is proposed in this paper. In this new method, both of the stroke pixels and the background pixels are assigned with different feature densities. In addition, the feature density is local and two-dimensional. It is more reasonable and more effective than other nonlinear shape normalization methods. Its effectiveness has been demonstrated by our experiments.
Key wordsoff-line handwritten Chinese character recognition preprocessing feature density feature density equalization nonlinear shape normalization
Unable to display preview. Download preview PDF.
- Yoshiyuki Yamashita et al, “Classification of Handprinted Kanji Characters by the Structured Segment Matching Method”, Pattern Recognition Letters 1 (1983) 475–479Google Scholar
- H.Yamada, T.Saito and K.Yamamoto, “Line Density Equalization—A Nonlinear Normalization for Correlation Method”, Trans. IECE, Vol.J67-D, No. 11, pp.1379–1383, 1984Google Scholar
- Jun Tsukumo, Haruhiko Tanaka, “Classification of Handprinted Chinese Characters Using Nonlinear Normalization and Correlation Methods”, Proc. of 9th ICPR, Rome, Itlay, Nov. 1988Google Scholar
- H.Yamada et al, “A Nonlinear Normalization Method for Handprinted Kanji Character Recognition—Line Density Equalization”, Pattern Recognition, Vol. 23, No.9, 1990Google Scholar
- S.W.Lee and J.S.Park, “Nonlinear Shape Normalization Methods for the Recognition of Large-Set Handwritten Characters”, Pattern Recognition, Vol. 27, No.7, pp.895–902, 1994Google Scholar
- Dayin Gou, “Research on Subject-Free Handprinted Chinese Character Recognition”, Ph.D. dissertation, Dept. of Electronic Engineering, Tsinghua University, Beijing, May 1995Google Scholar
- Dayin Gou, Xiaoqing Ding and Youshou Wu, A Handwritten Chinese Character Recognition Method Based on Image Shape Correction, Proc. of 1st National Conf. on Multimedia and Infomation Networks(CMIN'95), Beijing, Mar.1995, pp.254–259Google Scholar
- N.Sun et al, “A Handwritten Character Recognition System by Using Improved Directional Element Feature and Subspace Method”, IEICE Vol. J78-D-II, No.6, pp.922–930, June 1995 *** DIRECT SUPPORT *** A0008188 00007Google Scholar