魏湘辉,马少平. 基于凸包像素比特征的粘连汉字切分[J]. 中文信息学报, 2005, 19(1): 92-98.
WEI Xiang-hui , MA Shao-ping. Segmentation of Touching Chinese Character Based on Convex Hull Ratio Feature. , 2005, 19(1): 92-98.
Segmentation of Touching Chinese Character Based on Convex Hull Ratio Feature
WEI Xiang-hui , MA Shao-ping
1.Institute of software , CAS ,Beijing 10080 ,China ;2.State Key Laboratory of Intelligent Technology and Systems , Dept . of Computer Science , Tsinghua University Beijing 100084 ,China
Abstract：Accuracy of segmenting Chinese characters , especially touching characters , is essential for performance of a Chinese characters recognition system. The paper applied a background2thinning algorithm to segment two2touching Chinesecharacters that come from the dataset of four vaults. A newfeature called convex hull ratio was proposed for selection of the best segmentation path , as this feature exploits the property on the balance of Chinese charactersp structure. The experimental results show that segmentation accuracy improved consistently using the new feature when three different classifiers were experimented. And gaussian mixture model achieves the accuracy of 8816 %.
[1 ] Lin Yu Tseng , Rung Ching Chen. Segmenting handwritten Chinese characters based on heuristic merging of stroke bounding boxes and dynamic programming[J ] . Pattern Recognition Letters 1998 ,19 (10) : 963 - 973. [2 ] 1 N. W. Strathy , C. Y. Suen and A. Kryzak , Segmentation of handwritten digits using contour features[A] . Second Int. Conf . Document Anal. Recognition [C] 1993 ,577 - 580. [3 ] G. Congedo , G. Dimauro , S. Impedovo , and G. Pirlo. Segmentation of Numeric Strings[A] . Proc. of Third Int.Conf . on Document Analysis and Recognition[C] . Montreal :1995 ,1038 - 1041. [4 ] Yi2Kai Chen , Jhing2Fa Wang. Segmentation of Single2 or Multiple2 Touching Handwritten Numeral String Using Background and Foreground Analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2000 ,22(11) :1304 - 1317. [5 ] Shuyan Zhao , Zheru Chi , Penfei Shi , HongYan. Two2stage segmentation of unconstrained handwritten Chinese Characters[J ] . Pattern Recognition ,2003 ,36 (1) : 145 - 146. [ 6 ] ZhongkangLu , Zheru Chi , Wan2Chi Siu , Pengfei Shi. A background2thinning2based approach for separating and recognizing connected handwritten digit strings[J ] . Pattern Recognition , 1999 ,32 (6) : 921 - 933. [7 ] John C. Platt. Probabilistic outputs for supports vector machines and comparisons to regularized likelihood methods 96[ EB/ OL ] . http :/ / research. microsoft. com/ ～jplatt. [8 ] LIU Li , HEJialong. On the use of orthogonal GMMin speaker recognition [A] . 1999 IEEE International Conference on Acoustics Speech and Signal Proc [C] . 1999. 8452848. [9 ] R. C. Gonzalez and R. E. Woods , Digital Image Processing[M] . Boston : Addison2Wesley , 1992. [10 ] 陈强, 吕俊洋, 夏德深. 一种手写体大写金额串的分割新方法[J ] . 中文信息学报, 2004 , 18 (3) :66 -72. [11 ] 卢达, 谢铭培, 钱忆平, 浦炜. 一种基于骨架法形态分析的粘连字符图像分切方法[J ] . 中文信息学报, 1999 ,13 (2) :40 - 45.
2005, Vol. 19 
