汉字笔迹鉴别算法的研究
|
摘要
采用图像处理技术和模式识别理论研究了文本独立的计算机离线笔迹鉴别的方法,建立了一套能够反映笔迹特征的纹理参数体系,根据这些参数建立了笔迹鉴别的模式识别方法,为计算机笔迹鉴别打下理论和技术基础。
收集了60个人的笔迹样本,通过扫描仪转化为数字图像,建立了包含360(60×6)幅图像的笔迹鉴别样本库。对笔迹图像的预处理过程包括纸张背景颜色去除,狄度化、消噪和二值化,归一化等步骤。采用图像像素取色器去除背景颜色;分析了三种灰度化方法,确定采用加权平均法进行灰度化;研究了两种消噪方法,根据实验确定采用中值滤波法消除噪声;采用S.Watanabe方法进行二值化;归一化处理包括倾斜校正、去除标点符号、字符分割、字符尺寸归一化和文字拼接,分析了三种尺寸归一化方法,通过实验比较确定采用单边定界法。
分析了四种常用的纹理分析方法,确定采用Gabor变换进行笔迹纹理分析。研究了Gabor变换的特点和性质,并根据最优滤波器设计原则设计了Gabor滤波器。通过设置三种不同的Gabor滤波器参数获取了三套不同的笔迹纹理特征参数,其中第一套16个特征,第二套48个特征,第三套24个特征。
研究了不同核函数对支持向量机分类性能的影响。通过实验比较了k-近邻、BP神经网络和支持向量机的分类性能,针对支持向量机参数和核函数参数选择问题,采用了遗传算法进行参数优化,在给定的参数范围内得到了使支持向量机分类性能较好的参数。采用基于遗传算法优化参数的支持向量机(高斯径向机核函数)作为分类器对未知样本进行鉴别。
研究了模式识别中的特征选择方法,采用最近邻分类正确率作为特征选择的性能评价函数。通过实验比较了模拟退火法和遗传算法两种优化方法的搜索性能,确定采用基于最近邻分类器分类正确率——遗传算法的特征选择方法。
对比了3套特征参数在特征选择前后的分类结果。最终确定了表征笔迹纹理的参数体系和模式识别方法。最后对大样本情况下的笔迹鉴别做了探讨。
本研究根据笔迹纹理特征对离线文本独立的笔迹鉴别,其结果能够为计算机代替人进行笔迹鉴别提供一定的参考,丰富了图像处理领域关于笔迹分析和鉴别的方法。
In this study, image processing and pattern recognition theory of computer off-line handwriting text independent method is discussed to identify a set of texture features to reflect the parameters of the handwriting system, as well as the realization of these parameters in accordance with the pattern recognition method to identify the handwriting. The handwriting for the computer provides a theoretical basis for identification, and lays a solid foundation for the theory and technology.
60 samples of handwriting are collected and changed into digital images through a scanner, building a sample database of the handwriting identification containing 360 (60 x 6) images. The paper is divided into some steps such as pre-processing to remove background color, gray, de-noising, binarization and normalized. The background color is removed by the screen color device; three gray-scale methods are analyzed to determine the method using the weighted average of gray; two types of de-noising method are studied, according to the experiment to determine the use of median filter to eliminate noise; S. Watanabe methods are used for binary; normalized includes tip-tilt correction, removal of punctuation, character segmentation, character normalization size and the letter of Mosaic, and an analysis of three methods of size normalized by the experiment will be compared to determine the use of unilateral bound method.
Analyzing four common methods of texture analysis, Gabor transform is choosed to identify handwriting using texture analysis. Studying the characteristics and nature of Gabor transform, the Gabor filter is designed with the principles of optimal filter. By setting three different Gabor filter parameters the study obtains three different features of handwriting texture parameters, the first set of 16 features, the second set of 48 features, and the third set of 24 features.
The paper studies the different kernel function for SVM, and compares the classification performance of k-neighbor, BP neural network and SVM through the experiment. As to the parameters'selection for SVM and kernel function, the paper uses the genetic algorithm to optimize parameters, then gets better classification performance parameters of SVM in a given framework. This study identifies SVM (the Gaussian RBF kernel function) which is based on genetic algorithm to optimize its parameters as classifier to identify unknown samples.
The paper studies the methods of feature selection in pattern recognition, and uses nearest neighbor classification accuracy as the evaluation criteria for feature selection. Through comparing with the searching performance of the two kinds of optimization method of genetic algorithm and simulated annealing, the paper uses the feature selection methods based on neighbor classifier classification accuracy—genetic algorithm.
Comparing with three sets of characteristic parameters in the classification results before and after feature selection, the study ultimately sets the parameters for texture characterization of handwriting recognition systems and methods. Finally, the paper gives a brief discussion on identifying the handwriting in case of large samples.
The study identifies the off-line independent handwriting text based on texture feature The result can provide a powerful handwriting identification reference as the computer substitute for human, enriches the field of image processing on the handwriting analysis and identification method.
收集了60个人的笔迹样本,通过扫描仪转化为数字图像,建立了包含360(60×6)幅图像的笔迹鉴别样本库。对笔迹图像的预处理过程包括纸张背景颜色去除,狄度化、消噪和二值化,归一化等步骤。采用图像像素取色器去除背景颜色;分析了三种灰度化方法,确定采用加权平均法进行灰度化;研究了两种消噪方法,根据实验确定采用中值滤波法消除噪声;采用S.Watanabe方法进行二值化;归一化处理包括倾斜校正、去除标点符号、字符分割、字符尺寸归一化和文字拼接,分析了三种尺寸归一化方法,通过实验比较确定采用单边定界法。
分析了四种常用的纹理分析方法,确定采用Gabor变换进行笔迹纹理分析。研究了Gabor变换的特点和性质,并根据最优滤波器设计原则设计了Gabor滤波器。通过设置三种不同的Gabor滤波器参数获取了三套不同的笔迹纹理特征参数,其中第一套16个特征,第二套48个特征,第三套24个特征。
研究了不同核函数对支持向量机分类性能的影响。通过实验比较了k-近邻、BP神经网络和支持向量机的分类性能,针对支持向量机参数和核函数参数选择问题,采用了遗传算法进行参数优化,在给定的参数范围内得到了使支持向量机分类性能较好的参数。采用基于遗传算法优化参数的支持向量机(高斯径向机核函数)作为分类器对未知样本进行鉴别。
研究了模式识别中的特征选择方法,采用最近邻分类正确率作为特征选择的性能评价函数。通过实验比较了模拟退火法和遗传算法两种优化方法的搜索性能,确定采用基于最近邻分类器分类正确率——遗传算法的特征选择方法。
对比了3套特征参数在特征选择前后的分类结果。最终确定了表征笔迹纹理的参数体系和模式识别方法。最后对大样本情况下的笔迹鉴别做了探讨。
本研究根据笔迹纹理特征对离线文本独立的笔迹鉴别,其结果能够为计算机代替人进行笔迹鉴别提供一定的参考,丰富了图像处理领域关于笔迹分析和鉴别的方法。
In this study, image processing and pattern recognition theory of computer off-line handwriting text independent method is discussed to identify a set of texture features to reflect the parameters of the handwriting system, as well as the realization of these parameters in accordance with the pattern recognition method to identify the handwriting. The handwriting for the computer provides a theoretical basis for identification, and lays a solid foundation for the theory and technology.
60 samples of handwriting are collected and changed into digital images through a scanner, building a sample database of the handwriting identification containing 360 (60 x 6) images. The paper is divided into some steps such as pre-processing to remove background color, gray, de-noising, binarization and normalized. The background color is removed by the screen color device; three gray-scale methods are analyzed to determine the method using the weighted average of gray; two types of de-noising method are studied, according to the experiment to determine the use of median filter to eliminate noise; S. Watanabe methods are used for binary; normalized includes tip-tilt correction, removal of punctuation, character segmentation, character normalization size and the letter of Mosaic, and an analysis of three methods of size normalized by the experiment will be compared to determine the use of unilateral bound method.
Analyzing four common methods of texture analysis, Gabor transform is choosed to identify handwriting using texture analysis. Studying the characteristics and nature of Gabor transform, the Gabor filter is designed with the principles of optimal filter. By setting three different Gabor filter parameters the study obtains three different features of handwriting texture parameters, the first set of 16 features, the second set of 48 features, and the third set of 24 features.
The paper studies the different kernel function for SVM, and compares the classification performance of k-neighbor, BP neural network and SVM through the experiment. As to the parameters'selection for SVM and kernel function, the paper uses the genetic algorithm to optimize parameters, then gets better classification performance parameters of SVM in a given framework. This study identifies SVM (the Gaussian RBF kernel function) which is based on genetic algorithm to optimize its parameters as classifier to identify unknown samples.
The paper studies the methods of feature selection in pattern recognition, and uses nearest neighbor classification accuracy as the evaluation criteria for feature selection. Through comparing with the searching performance of the two kinds of optimization method of genetic algorithm and simulated annealing, the paper uses the feature selection methods based on neighbor classifier classification accuracy—genetic algorithm.
Comparing with three sets of characteristic parameters in the classification results before and after feature selection, the study ultimately sets the parameters for texture characterization of handwriting recognition systems and methods. Finally, the paper gives a brief discussion on identifying the handwriting in case of large samples.
The study identifies the off-line independent handwriting text based on texture feature The result can provide a powerful handwriting identification reference as the computer substitute for human, enriches the field of image processing on the handwriting analysis and identification method.
引文
[1]R.Lazarick.Multibiometric techniques and standards activities[A].39th Annual Technology [C],International Carnahan Conference on Security 2005:193-199.
[2]朱勇,谭铁牛,王蕴红.基于笔迹的身份鉴别[J].自动化学报,2001,27(2):229-234.
[3]R Plamond,G.Lorette.Automatic signature verification and writer identification-the state of art[J].Pattern Recognition,1989,22(2):107-131.
[4]F.Ramann,C.Vielhauer,R.Steinmetz.Biometric applications based on handwriting[A].IEEE International Conference on Multimedia Expo[C],2002,2:573-576.
[5]李莹.汉字笔迹鉴别的算法研究[D].山东大学硕士论文,2007.4.
[6]T.Wakahara,H.Murase,K.Odaka.On-line handwriting recognition[C].Proceedings of the IEEE,1992,80(7):1181-1194.
[7]N.Arica,F.T.Yarman-Vural.An overview of character recognition focusedon off-line handwriting[J].IEEE Trans.On Systems,Man and Cybernetics,2001,31(2):216-233.
[8]B.Azari.Automatic handwriting identification based on the external properties of the samples[J].IEEE Trans.Systems,Man and Sybernetics,1983,13(1):38-62.
[9]R.D.Naske.Writer recognition by prototype related deformation of handprinted Characters[A].Proc.6th ICPR[C],1982:819-822.
[10]I.Yoshimura,M.Yoshimura.Writer identification based on the arc pattern transformation [A].Proc.9th ICPR[C],1988:35-37.
[11]I.Yoshimura,M.Yoshimura.Off-line Writer Identification Using Ordinary Characters as the Object[J].Pattern Recognition,1991,24(9):909-915.
[12]S.Impedovo,et al.An Off-line Writer Identification System Based on Syntactic Approach [C].Proc.IWFHR,1990:53-61.
[13]W.Kuckuck.Writer recognition by spectral analysis[A].Int.Conf.Security through Science &Engineering[C],1980,1-3.
[14]B.Azari.Handwriting identification by means of run-length measurements[J].IEEE Trans.SMC,1977,7(12):878-881.
[15]W Kuckuck,B Rieger,K Steinker.Automatic writer recognition[A].Carnahan Conf.On Crime Countermeasures[C],1979,57-64.
[16]Amit Jain,Aditya Kamat.Personal Identification Based on Handwriting,the paper,2000
[17]M.Tuceryan,A.K.Jain.Texture analysis[A].The Handbook of Pattern Recognition and Computer Vision[C],C.H.Chen and L.F.Pau,Ed.River Edge,NJ:World Scientific,1998:207-248.
[18]S.N.Srihari,R.M.Bozinovic.Off-line cursive script word recognition[J].IEEE Trans. Pattern Analysis and Machine Intelligence,1989,11(1):68-83.
[19]清华大学.基于单个字符的统计笔迹鉴别和验证方法[P].中国:03109813.4,2005.5.4.
[20]郑建彬,杨亚莉.基于整数小波系数的笔迹图像鉴别方法研究[J].武汉理工大学学报(交通科学与工程版),2004,28(1):110-113.
[21]黄雅平,罗四维,陈恩义.基于独立分量分析的笔迹识别[J].中文信息学报,200317(4):52-58.
[22]欧贵文,肖国华.基于支持向量机的笔迹鉴别系统[J].中国图象图形学报A,2003,8(z1):551-554.
[23]杨磊,赵明旺,杨杰.基于能量解析的计算机笔迹特征提取[J].武汉科技大学学报(自然科学版),2006,29(1):79-82.
[24]贾永红.计算机图像处理与分析[M].武汉:武汉大学出版社,2001,166-163.
[25]黄端琼.多尺度纹理特征分析及其在遥感影像分类中的应用[D].福州大学,2006:9-10.
[26]容观澳.计算机图象处理[M].北京:清华大学出版社,2000:289.
[27]艾海舟.武勃等译.图像处理分析与机器视觉(第二版)[M].北京:人民邮电出版社,2003:448.
[28]徐光裕.计算机视觉[M].北京:清华大学第七届优秀讲义,2002:160-161.
[29]H.E.S.Said,T.N.Tan,K.D.Baker.Personal identification based on handwriting[J],patter recognition.2000,12(33):149-160.
[30]师宝山,张贵州.笔迹鉴别预处理算法的设计与实现[J].电子器件,2008.8:1357-1360.
[31]杨亚莉.基于纹理分析的笔迹鉴别方法研究[D].武汉理工大学,2004.5.
[32]何东健.数字图像处理技术[M].西安:西安电子科技大学出版社,2002.
[33]李弼程,彭天强,彭波等.智能图像处理技术[M].电子工业出版社,2004.
[34]刘成林,戴汝为,刘迎建.笔迹鉴别的字符预处理与匹配[J].中文信息学报,1997,10(3):50-57.
[35]赵锋,赵荣椿.纹理分割及特征提取方法综述[J].中国体视学与图像分析,1998,3(4):238-245.
[36]于海鹏,刘一星等.木材纹理的定量化算法探究[J].福建林学院学报,2005,25(2):157-162.
[37]钟克洪等.基于小波差分统计特性的纹理缺陷检测方法[J].系统工程与电子技术,2004,26(5):660-665.
[38]S.Arivazhagan,L.Ganesan.Texture Segmentation Using Wavelet Transform[J].Pattern Recognition Letters,2003,24(30):3197-3203.
[39]A.Bodnarova,M.Bennamoun,S.Latham.Optimal Gabor Filters for Textile Flaw Detection[J].Pattern Recognition,2002,35(29):2973-2291.
[40]D.Gabor.Theory of communications,J.Inst.Elec.Eng.1946,93:429-457.
[41]阮秋琦.数字图像处理学[M].北京:电子工业出版社,2001:126-127.
[42]G.H.Granlund.In search of a general and picture processing operator[J].Computer Graphics and Image Processing,1978,8(2):155-173.
[43]J.G Daugtnan.Two-dimensional spectral analysis of cortical receptive field profiles[J].Vision Research,1980,20(5):847-856.
[44]J.G Daugman.Uncertainty relation for resolution in space,spatial frequency,and orientation optimized by 2D visual cortical filters[J].Journal of the Optical Society of America,1985,2(7):1160-1169.
[45]J.G Daugtnan.Two-dimensional spectral analysis of cortical receptive field profiles[J].Vision Research,1980,20(5):847-856.
[46]T.S.Lee.Image representation using 2D Gabor wavelets[J].IEEETransactions on Pattern Analysis and Machine Intelligence,1996,18(10):959-971.
[47]A.Grossntann,J.Morlet.Decomposition of hardy functions into square integrable wavelets of constant shape[J].SIAM J.Math.Anal,1984,15:723-736.
[48]D.H.Liu,K.M.Lam,L.Shen.Optimal sampling of Gabor features for face Recognition [J].Pattern Recognition Leters,2004,25:267-276.
[49]曲宏山,郭小波,刘永平.改进Gabor滤波器在笔迹鉴别中的应用[J].计算机与数字工程,2008,8:146-149.
[50]B.Manjunath,W.Ma.Texture feature for browsing and retrieval of image data[J].IEEE Trans.on Pattern Anal and Machine Intel.1996,18(8):837-842.
[51]M.R.Turner.Textured discrimination by Gabor functions[J].Biological Cybernetics,1986,55(2-3):71-82.
[52]T.N.Tan.Texture edge detection by modeling visual cortical channels[J].Pattern Recogntion,1995,28(9):1283-1298.
[53]沈聪.基于改进的多通道Gabor小波变换的笔迹鉴别[D].北京工业大学硕士论文,2002.5.
[54]边肇祺,张学工.模式识别(第二版)[M].北京:清华大学出版社,2000.
[55]孙即祥.现代模式识别[M].长沙:国防科技大学出版社,2001.
[56]杨光正,吴岷,张晓莉.模式识别[M].合肥:中国科学技术大学出版社,2001.
[57]袁曾任.人工神经元网络及其应用[M].北京:清华大学出版社,1999:66.
[58]高隽.人工神经网络原理及仿真实例fM].北京:机械工业出版社,2003.
[59]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.
[60]Vapnik V,Levin E,Le C Y.Measuring the VC-dimension of a learning machine[J].Neural Computation,1994,6:851-876.
[61]Christopher J C.A tutorial on support vector machines for pattern recognition[J].Data Mining and Knowledge Discovery,1998,2:121-167.
[62]Cortes C,Vapnik V.Support-vector networks[J].Machine Learning,1995,20:273-297.
[63]Vapnik V N.The Nature of Statistical Learning Theory[M].New York:Springer-Verlag,1995.
[64]Vapnik V N.Statistical Learning Theory[M].New York:John Wiley & Sons,1998.
[65]唐发明.基于统计学习理论的支持向量机算法研究[D].武汉:华中科技大学博士学位论文,2005.
[66]Suyken J A K,Vandewalle J.Least squares support vector machine classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[67]朱家元,陈开陶,张恒喜.最小二乘支持向量机算法研究[J].计算机科学,2003,30(7):157-159.
[68]张鸿宾,孙广煜.Tabu搜索在特征选择中的应用[J].自动化学报,1999,25(4):457-466.
[69]王娟,慈林林,姚康泽.特征选择方法综述[J].计算机工程与科学,2005 27(12):68-71.
[70]李云.特征选择算法及其在基干内容图像检索中的应用研究[D].重庆:重庆大学博士论文,2005.
[71]Kirkpatrick S,Gelatt Jr C D,Vecchi M P.Optimization by simulated annealing[J].Science,1983,220:671-680.
[72]Van Laarhoven P J M,Aarts E H L.Simulated Anntaling Theory and Applications[M].Dordrecht:Reidel Publishing Company,1987.
[73]Aarts E,Korst J.Simulated Annealing and Boltzman Machines:A stochastic Approach to Combinatorial Optimization and Neural Computing[M].John Wiley & Sons,1989.
[74]Metropolis N,Rosenbluth A W,Rosenbluth M N,et al.Equations of state calculation by fast computing machines[J].The Journal of Chemical Physics,1953,21(3):1087-1092.
[75]Gber L.Simulated annealing practice versus theory[J].Mathematical and Computer Modeling,1993,18(1):29-57.
[76]王凌.智能优化算法及其应用[M].北京:清华大学出版社,2001:20,27-29.
[77]Metropolis N,Rosenbluth A W,Rosenbluth M N,et al.Equations of state calculations by fast computing machines[J].J Chem Phys,1953,21:1087-1091.
[78]李文勇,李泉永.基于模拟退火的全局优化算法[J].桂林电子工业学院学报,2001,21(2):33-37.
[79]康立山.非数值并行算法——模拟退火算法[M].北京:科学出版社,1994.
[80]Holland J H.Adaptation in Natural and Artificial Systems[M].Ann Arbor:Univ.of Michigan Press,1975.
[81]蔡自兴,徐光祐.人工智能及其应用(第三版)[M].北京:清华大学出版社,2004:152-166.
[82]韩炜,廖振鹏.关于遗传算法收敛性的注记[J].地震工程与工程振动.1999,19(4): 13-16.
[83]李敏强,寇纪淞,林丹等.遗传算法的基本理论与应用[M].北京:科学出版社,2002.
[84]雷英杰,张善文,李续武等.MATLAB遗传算法工具箱及其应用.西安:西安电子科技大学出版社,2005.
[85]吴晓涛,孙增圻.用遗传算法进行路径规划[J].清华大学学报(自然科学版),1995,35(5):14-19.
[86]周明,孙树栋.遗传算法原理及应用[M].长沙:国防科技大学出版社,2002:33-40.
[87]李敏强,寇纪淞,林丹等.遗传算法的基本理论与应用[M].北京:科学出版社,2002.
[88]王小平,曹立明.遗传算法——理论、应用与软件实现[M].西安:西安交通大学出版社,2002.
[89]王磊,潘进,焦李成.免疫算法[J].电子学报,2000,28(7):74-78.
[90]http://www.mathworks.cn.
[91]http://www.csie.ntu.edu.tw/～cjlin/.
[2]朱勇,谭铁牛,王蕴红.基于笔迹的身份鉴别[J].自动化学报,2001,27(2):229-234.
[3]R Plamond,G.Lorette.Automatic signature verification and writer identification-the state of art[J].Pattern Recognition,1989,22(2):107-131.
[4]F.Ramann,C.Vielhauer,R.Steinmetz.Biometric applications based on handwriting[A].IEEE International Conference on Multimedia Expo[C],2002,2:573-576.
[5]李莹.汉字笔迹鉴别的算法研究[D].山东大学硕士论文,2007.4.
[6]T.Wakahara,H.Murase,K.Odaka.On-line handwriting recognition[C].Proceedings of the IEEE,1992,80(7):1181-1194.
[7]N.Arica,F.T.Yarman-Vural.An overview of character recognition focusedon off-line handwriting[J].IEEE Trans.On Systems,Man and Cybernetics,2001,31(2):216-233.
[8]B.Azari.Automatic handwriting identification based on the external properties of the samples[J].IEEE Trans.Systems,Man and Sybernetics,1983,13(1):38-62.
[9]R.D.Naske.Writer recognition by prototype related deformation of handprinted Characters[A].Proc.6th ICPR[C],1982:819-822.
[10]I.Yoshimura,M.Yoshimura.Writer identification based on the arc pattern transformation [A].Proc.9th ICPR[C],1988:35-37.
[11]I.Yoshimura,M.Yoshimura.Off-line Writer Identification Using Ordinary Characters as the Object[J].Pattern Recognition,1991,24(9):909-915.
[12]S.Impedovo,et al.An Off-line Writer Identification System Based on Syntactic Approach [C].Proc.IWFHR,1990:53-61.
[13]W.Kuckuck.Writer recognition by spectral analysis[A].Int.Conf.Security through Science &Engineering[C],1980,1-3.
[14]B.Azari.Handwriting identification by means of run-length measurements[J].IEEE Trans.SMC,1977,7(12):878-881.
[15]W Kuckuck,B Rieger,K Steinker.Automatic writer recognition[A].Carnahan Conf.On Crime Countermeasures[C],1979,57-64.
[16]Amit Jain,Aditya Kamat.Personal Identification Based on Handwriting,the paper,2000
[17]M.Tuceryan,A.K.Jain.Texture analysis[A].The Handbook of Pattern Recognition and Computer Vision[C],C.H.Chen and L.F.Pau,Ed.River Edge,NJ:World Scientific,1998:207-248.
[18]S.N.Srihari,R.M.Bozinovic.Off-line cursive script word recognition[J].IEEE Trans. Pattern Analysis and Machine Intelligence,1989,11(1):68-83.
[19]清华大学.基于单个字符的统计笔迹鉴别和验证方法[P].中国:03109813.4,2005.5.4.
[20]郑建彬,杨亚莉.基于整数小波系数的笔迹图像鉴别方法研究[J].武汉理工大学学报(交通科学与工程版),2004,28(1):110-113.
[21]黄雅平,罗四维,陈恩义.基于独立分量分析的笔迹识别[J].中文信息学报,200317(4):52-58.
[22]欧贵文,肖国华.基于支持向量机的笔迹鉴别系统[J].中国图象图形学报A,2003,8(z1):551-554.
[23]杨磊,赵明旺,杨杰.基于能量解析的计算机笔迹特征提取[J].武汉科技大学学报(自然科学版),2006,29(1):79-82.
[24]贾永红.计算机图像处理与分析[M].武汉:武汉大学出版社,2001,166-163.
[25]黄端琼.多尺度纹理特征分析及其在遥感影像分类中的应用[D].福州大学,2006:9-10.
[26]容观澳.计算机图象处理[M].北京:清华大学出版社,2000:289.
[27]艾海舟.武勃等译.图像处理分析与机器视觉(第二版)[M].北京:人民邮电出版社,2003:448.
[28]徐光裕.计算机视觉[M].北京:清华大学第七届优秀讲义,2002:160-161.
[29]H.E.S.Said,T.N.Tan,K.D.Baker.Personal identification based on handwriting[J],patter recognition.2000,12(33):149-160.
[30]师宝山,张贵州.笔迹鉴别预处理算法的设计与实现[J].电子器件,2008.8:1357-1360.
[31]杨亚莉.基于纹理分析的笔迹鉴别方法研究[D].武汉理工大学,2004.5.
[32]何东健.数字图像处理技术[M].西安:西安电子科技大学出版社,2002.
[33]李弼程,彭天强,彭波等.智能图像处理技术[M].电子工业出版社,2004.
[34]刘成林,戴汝为,刘迎建.笔迹鉴别的字符预处理与匹配[J].中文信息学报,1997,10(3):50-57.
[35]赵锋,赵荣椿.纹理分割及特征提取方法综述[J].中国体视学与图像分析,1998,3(4):238-245.
[36]于海鹏,刘一星等.木材纹理的定量化算法探究[J].福建林学院学报,2005,25(2):157-162.
[37]钟克洪等.基于小波差分统计特性的纹理缺陷检测方法[J].系统工程与电子技术,2004,26(5):660-665.
[38]S.Arivazhagan,L.Ganesan.Texture Segmentation Using Wavelet Transform[J].Pattern Recognition Letters,2003,24(30):3197-3203.
[39]A.Bodnarova,M.Bennamoun,S.Latham.Optimal Gabor Filters for Textile Flaw Detection[J].Pattern Recognition,2002,35(29):2973-2291.
[40]D.Gabor.Theory of communications,J.Inst.Elec.Eng.1946,93:429-457.
[41]阮秋琦.数字图像处理学[M].北京:电子工业出版社,2001:126-127.
[42]G.H.Granlund.In search of a general and picture processing operator[J].Computer Graphics and Image Processing,1978,8(2):155-173.
[43]J.G Daugtnan.Two-dimensional spectral analysis of cortical receptive field profiles[J].Vision Research,1980,20(5):847-856.
[44]J.G Daugman.Uncertainty relation for resolution in space,spatial frequency,and orientation optimized by 2D visual cortical filters[J].Journal of the Optical Society of America,1985,2(7):1160-1169.
[45]J.G Daugtnan.Two-dimensional spectral analysis of cortical receptive field profiles[J].Vision Research,1980,20(5):847-856.
[46]T.S.Lee.Image representation using 2D Gabor wavelets[J].IEEETransactions on Pattern Analysis and Machine Intelligence,1996,18(10):959-971.
[47]A.Grossntann,J.Morlet.Decomposition of hardy functions into square integrable wavelets of constant shape[J].SIAM J.Math.Anal,1984,15:723-736.
[48]D.H.Liu,K.M.Lam,L.Shen.Optimal sampling of Gabor features for face Recognition [J].Pattern Recognition Leters,2004,25:267-276.
[49]曲宏山,郭小波,刘永平.改进Gabor滤波器在笔迹鉴别中的应用[J].计算机与数字工程,2008,8:146-149.
[50]B.Manjunath,W.Ma.Texture feature for browsing and retrieval of image data[J].IEEE Trans.on Pattern Anal and Machine Intel.1996,18(8):837-842.
[51]M.R.Turner.Textured discrimination by Gabor functions[J].Biological Cybernetics,1986,55(2-3):71-82.
[52]T.N.Tan.Texture edge detection by modeling visual cortical channels[J].Pattern Recogntion,1995,28(9):1283-1298.
[53]沈聪.基于改进的多通道Gabor小波变换的笔迹鉴别[D].北京工业大学硕士论文,2002.5.
[54]边肇祺,张学工.模式识别(第二版)[M].北京:清华大学出版社,2000.
[55]孙即祥.现代模式识别[M].长沙:国防科技大学出版社,2001.
[56]杨光正,吴岷,张晓莉.模式识别[M].合肥:中国科学技术大学出版社,2001.
[57]袁曾任.人工神经元网络及其应用[M].北京:清华大学出版社,1999:66.
[58]高隽.人工神经网络原理及仿真实例fM].北京:机械工业出版社,2003.
[59]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.
[60]Vapnik V,Levin E,Le C Y.Measuring the VC-dimension of a learning machine[J].Neural Computation,1994,6:851-876.
[61]Christopher J C.A tutorial on support vector machines for pattern recognition[J].Data Mining and Knowledge Discovery,1998,2:121-167.
[62]Cortes C,Vapnik V.Support-vector networks[J].Machine Learning,1995,20:273-297.
[63]Vapnik V N.The Nature of Statistical Learning Theory[M].New York:Springer-Verlag,1995.
[64]Vapnik V N.Statistical Learning Theory[M].New York:John Wiley & Sons,1998.
[65]唐发明.基于统计学习理论的支持向量机算法研究[D].武汉:华中科技大学博士学位论文,2005.
[66]Suyken J A K,Vandewalle J.Least squares support vector machine classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[67]朱家元,陈开陶,张恒喜.最小二乘支持向量机算法研究[J].计算机科学,2003,30(7):157-159.
[68]张鸿宾,孙广煜.Tabu搜索在特征选择中的应用[J].自动化学报,1999,25(4):457-466.
[69]王娟,慈林林,姚康泽.特征选择方法综述[J].计算机工程与科学,2005 27(12):68-71.
[70]李云.特征选择算法及其在基干内容图像检索中的应用研究[D].重庆:重庆大学博士论文,2005.
[71]Kirkpatrick S,Gelatt Jr C D,Vecchi M P.Optimization by simulated annealing[J].Science,1983,220:671-680.
[72]Van Laarhoven P J M,Aarts E H L.Simulated Anntaling Theory and Applications[M].Dordrecht:Reidel Publishing Company,1987.
[73]Aarts E,Korst J.Simulated Annealing and Boltzman Machines:A stochastic Approach to Combinatorial Optimization and Neural Computing[M].John Wiley & Sons,1989.
[74]Metropolis N,Rosenbluth A W,Rosenbluth M N,et al.Equations of state calculation by fast computing machines[J].The Journal of Chemical Physics,1953,21(3):1087-1092.
[75]Gber L.Simulated annealing practice versus theory[J].Mathematical and Computer Modeling,1993,18(1):29-57.
[76]王凌.智能优化算法及其应用[M].北京:清华大学出版社,2001:20,27-29.
[77]Metropolis N,Rosenbluth A W,Rosenbluth M N,et al.Equations of state calculations by fast computing machines[J].J Chem Phys,1953,21:1087-1091.
[78]李文勇,李泉永.基于模拟退火的全局优化算法[J].桂林电子工业学院学报,2001,21(2):33-37.
[79]康立山.非数值并行算法——模拟退火算法[M].北京:科学出版社,1994.
[80]Holland J H.Adaptation in Natural and Artificial Systems[M].Ann Arbor:Univ.of Michigan Press,1975.
[81]蔡自兴,徐光祐.人工智能及其应用(第三版)[M].北京:清华大学出版社,2004:152-166.
[82]韩炜,廖振鹏.关于遗传算法收敛性的注记[J].地震工程与工程振动.1999,19(4): 13-16.
[83]李敏强,寇纪淞,林丹等.遗传算法的基本理论与应用[M].北京:科学出版社,2002.
[84]雷英杰,张善文,李续武等.MATLAB遗传算法工具箱及其应用.西安:西安电子科技大学出版社,2005.
[85]吴晓涛,孙增圻.用遗传算法进行路径规划[J].清华大学学报(自然科学版),1995,35(5):14-19.
[86]周明,孙树栋.遗传算法原理及应用[M].长沙:国防科技大学出版社,2002:33-40.
[87]李敏强,寇纪淞,林丹等.遗传算法的基本理论与应用[M].北京:科学出版社,2002.
[88]王小平,曹立明.遗传算法——理论、应用与软件实现[M].西安:西安交通大学出版社,2002.
[89]王磊,潘进,焦李成.免疫算法[J].电子学报,2000,28(7):74-78.
[90]http://www.mathworks.cn.
[91]http://www.csie.ntu.edu.tw/～cjlin/.