一种基于特征提取的脱机手写汉字识别技术
|
摘要
本文的主要研究内容为：汉字识别的原理和方法，汉字识别前的预处理，脱机手写汉字的特征提取。
汉字识别的原理和方法介绍了汉字识别领域采用的一般方法和策略——基于数学特征的统计决策方法和基于结构特征的句法分析方法。
汉字图像的预处理包括对识别文稿进行平滑去噪、图像二值化、倾斜校正、行字切分以及归一化。
脱机手写汉字的特征提取在前两者的基础上，针对脱机手写汉字特点，找到了能充分反应手写汉字特点的三种特征并加以提取；同时提出了将汉字分解为部件来识别的观点。所提取的这些特征兼顾了提取方法的方便性和特征的稳定性，能有效地识别脱机手写汉字。
The main research content of this thesis include: the basic theory and method of Chinese character recognition, the pre-work of Chinese character recognition, the feature extraction of off-line handwritten Chinese character recognition.
The basic theory and method of Chinese character recognition introduces two basic thinking in the field of optic character recognition,which is statistical-decision algorithm based on math characteristic of character and structure-decomposition algorithm based on physical characteristic of character.
The pre-work of Chinese character recognition introduces five steps of optic character recognition,which is getting rid of noise, image binary, image incline rectify, image incise and image standardize.
The feature extraction of off-line handwritten Chinese character recognition which is based on the prior two, according to the features of handwritten Chinese characters, has found three features of handwritten Chinese characters and has extracted them. It puts forward a comprehensive viewpoint about the Chinese characters which should be separated first,then the character elements should be recognized.The features that we have found are convenient and stable,and they can recognize the handwritten Chinese characters effectively.
汉字识别的原理和方法介绍了汉字识别领域采用的一般方法和策略——基于数学特征的统计决策方法和基于结构特征的句法分析方法。
汉字图像的预处理包括对识别文稿进行平滑去噪、图像二值化、倾斜校正、行字切分以及归一化。
脱机手写汉字的特征提取在前两者的基础上，针对脱机手写汉字特点，找到了能充分反应手写汉字特点的三种特征并加以提取；同时提出了将汉字分解为部件来识别的观点。所提取的这些特征兼顾了提取方法的方便性和特征的稳定性，能有效地识别脱机手写汉字。
The main research content of this thesis include: the basic theory and method of Chinese character recognition, the pre-work of Chinese character recognition, the feature extraction of off-line handwritten Chinese character recognition.
The basic theory and method of Chinese character recognition introduces two basic thinking in the field of optic character recognition,which is statistical-decision algorithm based on math characteristic of character and structure-decomposition algorithm based on physical characteristic of character.
The pre-work of Chinese character recognition introduces five steps of optic character recognition,which is getting rid of noise, image binary, image incline rectify, image incise and image standardize.
The feature extraction of off-line handwritten Chinese character recognition which is based on the prior two, according to the features of handwritten Chinese characters, has found three features of handwritten Chinese characters and has extracted them. It puts forward a comprehensive viewpoint about the Chinese characters which should be separated first,then the character elements should be recognized.The features that we have found are convenient and stable,and they can recognize the handwritten Chinese characters effectively.
引文
[1]吴佑寿，丁晓青．汉字识别—原理、方法与实现．北京：高等教育出版社，1991．
[2]刘定一．文字、图形识别技术．北京：人民邮电出版社，1987．
[3]周昌乐．手写汉字的机器识别．北京：科学出版社，1992．
[4]吴佑寿．教电脑识字—浅谈汉字识别．北京：清华大学出版社，1991．
[5]丁晓青，郭繁夏．汉字识别技术的发展．北京：清华大学出版社，1993．
[6]张析中．汉字识别技术．北京：清华大学出版社，1992．
[7]胡家忠．计算机文字识别技术．北京：气象出版社，1993．
[8]Z.Zhao, M.Suters, H.Yan. Connected handwritten digit separation by optimal contour Partition. Proceedings of DICTA-93 Conference on Digital Image Computing Techniques and Applications, 1993,pp.786-793
[9]Bwonke H, Fliu San A. Chinese character recognition. Scientific World. 1989.
[10]N.W. Strathy, C.Y.Suen, A.Krzyzak. Segmentation of handwritten digits using contour features. Proceedings of the Second International Conference on Document Analysis and Recognition, 1993, pp.577-580
[11]M.Suters, H.Yan. Connected handwritten digits separation using external boundary curvature. J. Electron. Imaging, 1994, 3(3):251-256
[12]Dayin Gou, Xiaoqing Ding and Youshou Wu. A Handwritten Chinese Character Recognition Method Based on Image Shape Correction, Proc. of 1st National Conference on Multimedia and Informationn Networks(CMIN'95), Beijing, March 1995, pp.254-259
[13]Donggang. Yu, Hong Yan. Separation of touching handwritten multi-numeral strings based onMorphological structural features. Pattern Recognition, 2001, 34:587-599
[14]Yi-Kai Chen, Jhing-Fa Wang. Segmentation of Single-or Multiple-Touching Handwritten Transaction Numeral String Using Background and on Pattern Analysis and Machine Intelligence, Foreground Analysis. IEEE 2000, 220(11):1304-1317
[15]Shunji Mori, C.Y.Suen and Kazuhiko Yamamoto, Historical review of OCR research and development, Proceedinds of the IEEE, Vol.80, No.7, 1992, pp.1029-1058
[16]Wang L, Pavlidis T. Direct grayscale extraction of features for character recognition. IEEE Trans Pattern Anal Mach Intell, 1993, 15:1053-1069
[17]杨承磊，孟祥旭．一种新的快速细化算法的设计与实现．工程图学学报，1998，3：87-93
[18]韩燮，张永梅，刘幼立．汉字识别的方法及Rosen细化算法的改进．华北工学院学报，1997．18(1)
[19]李存华．基于轮廓投影方法的文本图像偏斜纠正．中国图像图形学报，2001，10
[20]高彤，姜华，吕民．基于模板匹配的手写体字符识别方法．哈尔滨工业大学学报，1999， 31(1)：104-106
[21]娄震，胡钟山，胡静宇等．基于轮廓分段特征的手写体阿拉伯数字识别．计算机学报，22(10)：1065-1073
[22]沈会良，李志能．基于矩和小波变换的数字、字母字符识别研究．2000，5(A)(3)：249-252
[23]李莉，舒文豪．手写体汉字识别粗分类方法研究．模式识别与人工智能，1990，(2)
[24]王林泉．关于手写体汉字识别的研究．计算机研究与发展，1989，(5)
[25]Seong-Whan Lee and Jeong-Seon Park. Nonlinear Shape Normalization Methods For The Recognition of Large-set Handwritten Characters, Pattern Recognition, Vol. 27, pp. 895-902, 1994
[26]Hiromitsu Yamaa, Kazuhiko Yaxnamoto and Taiichi Saito. A Nonlinear Normalisation Method for Handprinted Kanji Character Recognition-Line Density Equalization, Pattern Recognition, vol. 23, No. 9, pp. 1023-1029,1990
[27]Y. Yamashita, K. Higuchi, Y. Yamada and Y. Haga. Classification of Handprinted Kanji Characters by the Structured Segment Matching Method, Pattern Recognition Letters, volum 1, numbers 5, 6, pp. 475-479, 1983
[28]SUENC Y, NADAL C, LEGAULT R, et al. Computer recognition of unconstrained handwritten numerals, prodeedings of the IEEE, 1992, 20(7),1162-1180
[29]HUANG Y S, SUEN C Y. A method of combining multiple experts for the recognition of unconstrained handwritten numerals. IEEE Trans, 1995, PAMI(17):90-94
[30]CAO J, AHAMADI M, SHRIDHAR M. Recogntion of Handwritten Numerals with Mutiple Feature and Mutistage classifier. Pattern Recognition, 1995, 28(2):153-160
[31]Tze Fen Li, and Shiaw Shian Yu. Handpiinted Chinese character recognition using the probability distribution feature, International Journal of Pattern Recognition and Artificial Intelligence, Vol. 8, No. 5, 1994,1241-1258
[2]刘定一．文字、图形识别技术．北京：人民邮电出版社，1987．
[3]周昌乐．手写汉字的机器识别．北京：科学出版社，1992．
[4]吴佑寿．教电脑识字—浅谈汉字识别．北京：清华大学出版社，1991．
[5]丁晓青，郭繁夏．汉字识别技术的发展．北京：清华大学出版社，1993．
[6]张析中．汉字识别技术．北京：清华大学出版社，1992．
[7]胡家忠．计算机文字识别技术．北京：气象出版社，1993．
[8]Z.Zhao, M.Suters, H.Yan. Connected handwritten digit separation by optimal contour Partition. Proceedings of DICTA-93 Conference on Digital Image Computing Techniques and Applications, 1993,pp.786-793
[9]Bwonke H, Fliu San A. Chinese character recognition. Scientific World. 1989.
[10]N.W. Strathy, C.Y.Suen, A.Krzyzak. Segmentation of handwritten digits using contour features. Proceedings of the Second International Conference on Document Analysis and Recognition, 1993, pp.577-580
[11]M.Suters, H.Yan. Connected handwritten digits separation using external boundary curvature. J. Electron. Imaging, 1994, 3(3):251-256
[12]Dayin Gou, Xiaoqing Ding and Youshou Wu. A Handwritten Chinese Character Recognition Method Based on Image Shape Correction, Proc. of 1st National Conference on Multimedia and Informationn Networks(CMIN'95), Beijing, March 1995, pp.254-259
[13]Donggang. Yu, Hong Yan. Separation of touching handwritten multi-numeral strings based onMorphological structural features. Pattern Recognition, 2001, 34:587-599
[14]Yi-Kai Chen, Jhing-Fa Wang. Segmentation of Single-or Multiple-Touching Handwritten Transaction Numeral String Using Background and on Pattern Analysis and Machine Intelligence, Foreground Analysis. IEEE 2000, 220(11):1304-1317
[15]Shunji Mori, C.Y.Suen and Kazuhiko Yamamoto, Historical review of OCR research and development, Proceedinds of the IEEE, Vol.80, No.7, 1992, pp.1029-1058
[16]Wang L, Pavlidis T. Direct grayscale extraction of features for character recognition. IEEE Trans Pattern Anal Mach Intell, 1993, 15:1053-1069
[17]杨承磊，孟祥旭．一种新的快速细化算法的设计与实现．工程图学学报，1998，3：87-93
[18]韩燮，张永梅，刘幼立．汉字识别的方法及Rosen细化算法的改进．华北工学院学报，1997．18(1)
[19]李存华．基于轮廓投影方法的文本图像偏斜纠正．中国图像图形学报，2001，10
[20]高彤，姜华，吕民．基于模板匹配的手写体字符识别方法．哈尔滨工业大学学报，1999， 31(1)：104-106
[21]娄震，胡钟山，胡静宇等．基于轮廓分段特征的手写体阿拉伯数字识别．计算机学报，22(10)：1065-1073
[22]沈会良，李志能．基于矩和小波变换的数字、字母字符识别研究．2000，5(A)(3)：249-252
[23]李莉，舒文豪．手写体汉字识别粗分类方法研究．模式识别与人工智能，1990，(2)
[24]王林泉．关于手写体汉字识别的研究．计算机研究与发展，1989，(5)
[25]Seong-Whan Lee and Jeong-Seon Park. Nonlinear Shape Normalization Methods For The Recognition of Large-set Handwritten Characters, Pattern Recognition, Vol. 27, pp. 895-902, 1994
[26]Hiromitsu Yamaa, Kazuhiko Yaxnamoto and Taiichi Saito. A Nonlinear Normalisation Method for Handprinted Kanji Character Recognition-Line Density Equalization, Pattern Recognition, vol. 23, No. 9, pp. 1023-1029,1990
[27]Y. Yamashita, K. Higuchi, Y. Yamada and Y. Haga. Classification of Handprinted Kanji Characters by the Structured Segment Matching Method, Pattern Recognition Letters, volum 1, numbers 5, 6, pp. 475-479, 1983
[28]SUENC Y, NADAL C, LEGAULT R, et al. Computer recognition of unconstrained handwritten numerals, prodeedings of the IEEE, 1992, 20(7),1162-1180
[29]HUANG Y S, SUEN C Y. A method of combining multiple experts for the recognition of unconstrained handwritten numerals. IEEE Trans, 1995, PAMI(17):90-94
[30]CAO J, AHAMADI M, SHRIDHAR M. Recogntion of Handwritten Numerals with Mutiple Feature and Mutistage classifier. Pattern Recognition, 1995, 28(2):153-160
[31]Tze Fen Li, and Shiaw Shian Yu. Handpiinted Chinese character recognition using the probability distribution feature, International Journal of Pattern Recognition and Artificial Intelligence, Vol. 8, No. 5, 1994,1241-1258