基于GSLPP特征提取算法和多分类器融合的人脸识别方法研究
|
摘要
人脸识别方法和关键技术是当前模式识别和计算机视觉领域的一个研究热点。人脸识别的步骤主要包括人脸检测、特征提取和特征分类。本文主要对人脸特征提取和特征分类进行了深入的探讨和研究,提出了基于Gabor小波变换的有监督局部保持投影算法(Based on Gabor Feature Supervised locality preservingprojection,GSLPP)并应用于人脸特征提取,及最近邻特征线分类器与支持向量机分类器二级融合的分类方法应用于特征分类,本文的主要工作如下:
(1)在人脸识别的特征提取阶段,本文采用了基于Gabor小波的一种有监督的LPP算法,该方法先用Gabor小波对人脸图像进行滤波处理,然后直接对每个LPP基向量进行线性判别分析,选择最具判别力的基向量来构造子空间。Gabor滤波器对于图像的亮度和对比度变化以及人脸姿态变化有较强的鲁棒性,并且它表达的是对人脸识别最为有用的局部特征。基于以上优点本文先采用Gabor小波对人脸图像进行滤波处理。提取人脸Gabor特征通常要对一幅图像进行四十次滤波处理,这会使图像的维数增加,从而影响识别的效率,为了提高识别的速度,必须采取有效的降维方法来降低数据的维数。本文采用有监督的局部保持投影算法提取经Gabor小波滤波后的人脸图像的主要特征。局部保持投影是一种新的子空间分析方法,它是非线性方法Laplacian Eigenmap的线性近似,既解决了PCA等传统线性方法难以保持原始数据非线性结构的缺点,又解决了非线性方法难以获得新样本点低维投影的缺点。但LPP算法是一种无监督的学习方法,当人脸图像的光照、姿态、表情发生变化时,LPP的识别率会迅速下降。本文提出的一种有监督的LPP算法,因其直接对每个LPP基向量进行线性判别分析,选择最具判别力的基向量来构造子空间,有效的提取了人脸图像的分类特征,实验证实其在特征提取方面具有较好的效果。
(2)在人脸识别的人脸特征分类阶段,本文提出最近邻特征线分类器与支持向量机分类器以串联的方式构建一个二级融合的分类器。如果将多种不同的分类器以某种方式进行组合,就有可能在总体上取得比单一分类器更好的分类效果。基于此观点,本文提出了一种将最近邻特征线分类器与支持向量机分类器进行二级融合的分类器设计方法。利用该融合分类器分类时,先采用最近邻特征线分类器进行第一级分类,若结果大于设定的阈值,则拒识,否则,转入后级分类器,用支持向量机分类器进行精确分类。该多分类器融合方法不仅充分利用了支持向量机分类器识别率高和最近邻特征线分类器速度快的优点,而且还利用最近邻特征线分类器的结果指导支持向量机分类器的训练和测试,从总体上提高了分类的精度。
应用上述方法,结合FERET人脸图像库和JDL人脸数据库进行实验,结果表明本论文提出的人脸识别方法与其它(PCA,LDA等)识别方法相比具有较好的人脸识别效果。
Face recognition is a focus of the research in the field of Pattern Recognition and Computer Vision which mainly includes face detection, facial feature extraction and facial feature classification. This paper discuss mainly on facial feature extraction and facial feature classification. This paper put forward a facial feature extraction algorithm which base on Gabor wavelet feature using supervised locality preserving projection, and a multi-classifiers fusion method which combine nearest feature line classifier and support vector machine classifier. The main contents of the paper can be noted as following:
(1) At the facial feature extraction stage, this paper proposes one facial feature extraction algorithm which base on Gabor wavelet feature using supervised locality preserving projection and is abbreviated GSLPP. Due to the good characteristic of human face image's Gabor wavelet feature, face recognition technology base on Gabor wavelet feature is a very popular method. However, human face image's Gabor filter processing will increase the dimension of data. In order to reduce the dimension of data, this paper use supervised locality preserving projection algorithm to reduce the dimension of data. Locality preserving projection algorithm is a new subspace analysis method, which can preserve the most important part for face recognition. The method was successfully applied to face recognition in controlled environment. But the locality preserving projection algorithm is a non-supervised study method. When the human face image's illumination, posture and expression changed, the recognition rate will decrease dramatically. For improving the recognition rate, this paper put forward a supervised locality preserving projection algorithm.
(2) Based on the view that combining some kinds of classifier by some way maybe obtain higher recognition rate than a single one, this paper put forward a multi-classifiers fusion method which combine nearest feature line classifier and support vector machine classifier. The step of the multi-classifiers fusion method as following: firstly, using the nearest feature line classifier to pre-stage classify, if the result bigger than the defined threshold, then rejecting recognition, else pass to SVM classifier to classify. The multi-classifiers fusion method not only makes full use of the advantage which is high recognition rate of SVM classifier and high recognition speed of the nearest feature line classifier,but also utilizes the classification result of the nearest feature line classifier to guide the training and classification of SVM classifier.
At last, the paper test this human face recognition method by some experiments on FERET Face Database and JDL-A Face Database. Those experiments illustrate that the method proposed by this paper is more effective than other methods, such as PCA、LDA and so on.
(1)在人脸识别的特征提取阶段,本文采用了基于Gabor小波的一种有监督的LPP算法,该方法先用Gabor小波对人脸图像进行滤波处理,然后直接对每个LPP基向量进行线性判别分析,选择最具判别力的基向量来构造子空间。Gabor滤波器对于图像的亮度和对比度变化以及人脸姿态变化有较强的鲁棒性,并且它表达的是对人脸识别最为有用的局部特征。基于以上优点本文先采用Gabor小波对人脸图像进行滤波处理。提取人脸Gabor特征通常要对一幅图像进行四十次滤波处理,这会使图像的维数增加,从而影响识别的效率,为了提高识别的速度,必须采取有效的降维方法来降低数据的维数。本文采用有监督的局部保持投影算法提取经Gabor小波滤波后的人脸图像的主要特征。局部保持投影是一种新的子空间分析方法,它是非线性方法Laplacian Eigenmap的线性近似,既解决了PCA等传统线性方法难以保持原始数据非线性结构的缺点,又解决了非线性方法难以获得新样本点低维投影的缺点。但LPP算法是一种无监督的学习方法,当人脸图像的光照、姿态、表情发生变化时,LPP的识别率会迅速下降。本文提出的一种有监督的LPP算法,因其直接对每个LPP基向量进行线性判别分析,选择最具判别力的基向量来构造子空间,有效的提取了人脸图像的分类特征,实验证实其在特征提取方面具有较好的效果。
(2)在人脸识别的人脸特征分类阶段,本文提出最近邻特征线分类器与支持向量机分类器以串联的方式构建一个二级融合的分类器。如果将多种不同的分类器以某种方式进行组合,就有可能在总体上取得比单一分类器更好的分类效果。基于此观点,本文提出了一种将最近邻特征线分类器与支持向量机分类器进行二级融合的分类器设计方法。利用该融合分类器分类时,先采用最近邻特征线分类器进行第一级分类,若结果大于设定的阈值,则拒识,否则,转入后级分类器,用支持向量机分类器进行精确分类。该多分类器融合方法不仅充分利用了支持向量机分类器识别率高和最近邻特征线分类器速度快的优点,而且还利用最近邻特征线分类器的结果指导支持向量机分类器的训练和测试,从总体上提高了分类的精度。
应用上述方法,结合FERET人脸图像库和JDL人脸数据库进行实验,结果表明本论文提出的人脸识别方法与其它(PCA,LDA等)识别方法相比具有较好的人脸识别效果。
Face recognition is a focus of the research in the field of Pattern Recognition and Computer Vision which mainly includes face detection, facial feature extraction and facial feature classification. This paper discuss mainly on facial feature extraction and facial feature classification. This paper put forward a facial feature extraction algorithm which base on Gabor wavelet feature using supervised locality preserving projection, and a multi-classifiers fusion method which combine nearest feature line classifier and support vector machine classifier. The main contents of the paper can be noted as following:
(1) At the facial feature extraction stage, this paper proposes one facial feature extraction algorithm which base on Gabor wavelet feature using supervised locality preserving projection and is abbreviated GSLPP. Due to the good characteristic of human face image's Gabor wavelet feature, face recognition technology base on Gabor wavelet feature is a very popular method. However, human face image's Gabor filter processing will increase the dimension of data. In order to reduce the dimension of data, this paper use supervised locality preserving projection algorithm to reduce the dimension of data. Locality preserving projection algorithm is a new subspace analysis method, which can preserve the most important part for face recognition. The method was successfully applied to face recognition in controlled environment. But the locality preserving projection algorithm is a non-supervised study method. When the human face image's illumination, posture and expression changed, the recognition rate will decrease dramatically. For improving the recognition rate, this paper put forward a supervised locality preserving projection algorithm.
(2) Based on the view that combining some kinds of classifier by some way maybe obtain higher recognition rate than a single one, this paper put forward a multi-classifiers fusion method which combine nearest feature line classifier and support vector machine classifier. The step of the multi-classifiers fusion method as following: firstly, using the nearest feature line classifier to pre-stage classify, if the result bigger than the defined threshold, then rejecting recognition, else pass to SVM classifier to classify. The multi-classifiers fusion method not only makes full use of the advantage which is high recognition rate of SVM classifier and high recognition speed of the nearest feature line classifier,but also utilizes the classification result of the nearest feature line classifier to guide the training and classification of SVM classifier.
At last, the paper test this human face recognition method by some experiments on FERET Face Database and JDL-A Face Database. Those experiments illustrate that the method proposed by this paper is more effective than other methods, such as PCA、LDA and so on.
引文
[1]Woodward.J.D.Biometrics.Privacy's foe or friend? IEEE Proc.1997,85(9):1480-1492.
[2]R.Chellappa,C.L.Wilson and S.Sirohey.Human and machine recognition of faces:A survey[J].IEEE Proc.1995,83(5):705-740.
[3]A.Samal and P.A.Iyengar.Automatic recognition and analysis of human faces and facial expressions:A survey[J].Pattern Recognition.1992,25(1):65-77.
[4]W.Zhao,R.Chellappa,P.J.Philips and A.Rosenfel.Face Recognition:A Literature Survey [J].ACM computring Surveys.2003,35(4):399-458.
[5]贾云得.机器视觉[M].1999科学出版社.
[6]Vapnik V N.An Overview of Statistical Learning Theory[J].IEEE Trans.On Neural Networks.1999,10(5):988-999.
[7]周激流,张哗.人脸识别理论研究进展[J].计算机辅助设计与图像学报.1999,11(2):23-29.
[8]Erik.Hjellnas,Boon.KeeLow.Face Detcetion:ASurvey[J].Computer Vision and Image understanding.2001,83:236-274.
[9]Y.Amit,D.Geman,B.Jedynak.Efficient focusing and face detection[J].Face Recognition.1998.
[10]梁路宏,艾海舟,徐光佑.人脸检测研究综述[J].计算机学报.2002.
[11]D.Demers,G.W.Cottrell.Nonlinear Dimensionality Reduetion[A].Advances in Neural Information Processing System[C],San Mateo,CA:Morgan Kaufinann.1993,580-587.
[12]Huang C L,Chen C W.Human facial feature extraction for face interpretation and Recognition[J].Pattern Recognition.1992,25(2):1435-1444.
[13]Kass M,Witkin A,Terzopoulos D.Snakes.Active contour models[J].International Journal of Computer Vision.1988,1(4):321-331.
[14]Sirovich L,Kirby M.Application of Karhunen-Loeve procedure for the characterization of human faces[J].IEEE Trans.PAMI,1990,3(1):71-79.
[15]Turk M A,Pentland A P.Eigenfaces for recognition[J].Journal of Cognitive Neuroscience,1991,3(1):71-86.
[16]Paul D.A neural network for facial feature location[J].UC Berkeley CS283 Project Report.1992,10:1-9.
[17]刘小军.人脸识别技术研究[D].中国科学院.2001.
[18]Lee S Y,Ham Y K,Park R H.Recognition of human front faces using knowledge-based feature extraction and neuro-fuzzy algorithm[J].Pattern Recognition,1996,29(11):1863-1879.
[19]Valentin D,Abdi H,O'Toole A,Cottrell G W.Connectionist models of face processing:A survey.Pattern Recognition.1994,27(9):1209-1230.
[20]Intrator N,Reisfeld D,Yeshurun Y.Face recognition using a hybrid supervised/unsupervised neural network[J].Pattern Recognition Letters.1996,17(1):67-76.
[21]Sam T,Roweis and Saul K L.Nonlinear dimensionality reduction by locally linear embedding[J].Science.2000,290(5500):2323-2326.
[22]He X and Niyogi.Locality preserving projections[C].Proceedings of Advances In Neural Information Processing Systems 16,MA:Cambridge,MIT Press.2004,153-160.
[23]Zhao Haitao,Sun Shaoyuan,Jing Zhongliang,and Jingyu Yang.Local structure based supervised feature extraction[J].Pattern Recognition.2006,39(881):1546-1550.
[24]M.Lades,J.C.Vorbruggen,J.Buhmann,J.Lange.Distortion invariant object recognition in the dynamic link architecture[J].Transcation Computer.1993,42:300-311.
[25]C.Liu and H.Wechsler.Gabor feature based classification using the enhanced fisher linear discriminant model for face recognition[J].Transaction on Image Processing.2002,11(4):467-476.
[26]Abhijit S.Pandya,Robert B.Macy著.徐勇,荆涛等译.神经网络模式识别及其实现[M].北京:电子工业出版社,1999:153-178.
[27]Gutta S,Huang J,Singh D.Benchmark.Studies on Face Recognition.Proceedings of International Workshop on Automatic Face and Gesture Recognition.IWAFGR 95,1995:227-231.
[28]S.Z.L,J.Lu.Face recognition using the nearest feature line method[J].Transactions on Neural Networks.11999,10(2):439-443.
[29]Burges C JC.A tutorial on support vector machines for pattern recognition[J].Data Mining and Knowledge Discovery.1998,2(2):57-66.
[30]SehoLkopf,Burges.VapnikV.Extracting Support Data for a Given Task[A]Proc.Of 1~(st) International Conference on Knowledge Discovery and Data Mining[C],AAAI,1995.252-257
[31]Vavnik V.The Nature of Statistical Learning Theory(Zndedition.)[M].New York,2000.
[32]Boser B,Guyonl,VapnikV.A Training Algorithm for Optimal Margin classifiers [A].Proceedings of Annual Workshop on Computational Learning Theory[C].Canada,1992.
[33]J.A.K.Suykens,J.Vandewalle.Last Squares Support Vector machine Classifiers[J].Neural processing Letters.1999,9(3):293-300.
[34]PeterAndras.The Equivalence of Support Machine and Regularization Neural Networks[J].Neural processing Letters.2002,15(2):97-104.
[35]刘江华,程君实,陈佳品.支持向量机训练算法综述.信息与控制.2002
[36]Chih WeiHsu.A companies on of methods for multi-class support vector machines[J].Trans Neural Networks.2002,13(2):415-425.
[37]U.KreBel.Pair wise classification and support vector machines[J].Advances in Kernel Methods:Support vector Leaning.1999,255-268.
[38]潘翔.基于多分类器集成的模式识别研究[D].2002,浙江工业大学.
[39]Killter J,Hatef M,Duin R P Wet al.On Combining Classifiers[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.1998,20(3):226-239.
[40]叶俊勇,汪同庆.基于支持向量机的人脸检测算法[J].计算机工程.2003,29(2):23-25.
[41]P.J.Phillips,H.Wechsler,J.Huang,P.Rauss,"The FERET database and evaluation procedure for face recognition algorithms," Image and Vision Computing J,Vol.16,No.5,pp.295-306,1998.
[42]P.J.Phillips,H.Moon,S.A.Rizvi,P.J.Rauss,"The FERET Evaluation Methodology for Face Recognition Algorithms," IEEE Trans.Pattern Analysis and Machine Intelligence,Vol.22,pp.1090-1104,2000.
[43]Dang-Hui Liu,Kin-Man Lain,Lan-Sun Shen.Optimal Sampling of Gabor Features for Face Recognition[J].Pattern Recognition Letters,2004,25:267-276.
[44]K.Jonsson,J.Matas,J.Kittler,Y.P.Li.Learning Support Vectors for Face Verification and Recontion[C].Proceeding of the 4th Int.Conf.on Auto.Face and GestureRecog:208-213,Grenoble,France,2000.3.
[45]Schapire,R.The Strength of Weak Leanability[J].Machine Leaning,1990,5(2):197-227.
[46]PengYang,Shiguang Shan,Wen Gao,Stan Z.Li,Dong Zhang.Face Recognition Using Ada-Boosted Gabor Features[J].Proc.Of Int.Conf.on Auto.Face and Gesture recognition,2004.
[47]Qijun Zhao,Hongtao Lu,and David Zhang.A Fast Evolutionary Pursuit Algorithm based on Linearly Combining Vectors[J].Pattern Recognition,2006,39:310-312.
[48]W.S.Zheng,J.H.Lal,and P.C.Yuen,GA-Fisher.A new LDA-based face recognition algorithm with selection of principal components[J].IEEE Transaction on Systems,Man,and Cybernetics-Part B,2005,5(35):1065-1078.
[49]D.Goldberg.Genetic Algorithms in Search,Optimization,and Machine Learning[M].Adison-Wesley,1989.
[50]R.Brunelli,T.Poggio.Face Recognition:Features vs.Templates[J].IEEE Trans.On PAMI,1993,15(10):1042-1052.
[51]B.Moghaddam.Principal Manifolds and Probabilistic subspaces for visual recognition[J].IEEE Trans.PAMI,2002,24(6) 780-788.
[52]B.Moghaddam and A.Pentland,Probabilistic Visual Learning for Object Representation[J].IEEE Trans.On PAMI,1997,20(7):696-710.
[53]L.Back,B.A.Draper,J.R.Beveridge and Kai She.PCA vs.ICA:A comparison on the FERET data set[C].Int'l Conf.on Computer Vision,Pattern Recognition and Image Processing,2002.
[54]P.Common.Independent Component Analysis-A New Concept for Signal Processing,1994,36:287-314.
[55]Yang,Miang-Hsuan,Kriegman,David J.,Ahuja,Narendra.Dectecting faces in images:A survey,Source[J].IEEE Tran.On PAMI,2002,24:34-58.
[56]Q.J.Zhao and H.T.Lu.GA-Driven LDA in KPCA Space for Facial Expression Recognition[J].Lecture Notes on Computer Science,2005,3611:8-36.
[57]Ming.Hsuan Yang.Recent Advances in Face Detection[R].IEEE.ICPR, 2004,Tutorial,Cambridge,UK.
[58]Chen,S.,Zhang,D.,and ZhouZ.-H.A new face recognition methodbased on SVD perturbation for single example image per person[C].Appl.Math.Comput,2005.
[59]徐科军,于翠欣.基于DSP的柯氏质量流量计信号处理系统[J].仪器仪表学报,2002,23(2):170-177.
[60]金忠,杨静宇,陆建峰.一种具有统计不相关的最佳鉴别矢量集[J].计算机学报,1999,22(10).
[61]边肇祺,张学工.模式识别(第二版)[M].北京:清华大学出版社,2000.
[62]梁路宏,等.基于多关联模板匹配的人脸检测[J].软件学报,2001,12(1):92-102.
[63]祝磊,朱善安.基于二维广义主成分分析的人脸识别[J].浙江大学学报,2007,41(2):264-267.
[64]陈伏兵,陈秀宏,高秀梅,等,二维主成分分析方法的推广及其在人脸识别中的应用[J].计算机应用,2005,25(8):1767-1770.
[65]王跃明.表情不变的三维人脸识别研究[D].浙江大学计算机学院博士学位论文,2007.
[66]刘向东.全自动三维人脸识别框架[D].吉林大学硕士学位论文,2007.
[67]赖剑煌,颜鑫弘,邓东皋.完善频谱脸人脸识别的分类器设计[J].中国图像图形学报,2002,7(5):466-471.
[68]赖剑煌,阮邦志,冯国灿.频谱脸.一种基于小波变换和Fourier变换的人脸识别新方法[J].中国图像图形学报,1999,4(10):811-817.
[69]杨琼,丁晓青.对称主分量分析及其在人脸识别中的应用[J].计算机学报,2003,26(9):1146-1151.
[70]刘青山,卢汉清,马颂德.综述人脸识别中的子空间方法[J].自动化学报,2003,29(6):900-911.
[71]彭岩,张道强.半监督典型相关分析算法[J].软件学报,2008,19(11):2822-2832.
[72]唐亮,熊蓉,褚健.单样本条件下权重模块2DPCA人脸识别[J].中国图象图形学报,2008,13(12):2307-2313.
[73]张莹,王耀南.基于Gabor滤波器包络的人脸识别算法[J].中国图象图形学报,2008,13(12):2314-2320.
[74]柴秀娟,山世光,卿来云,陈熙霖,高文.基于3D人脸重建的光照、姿态不变人脸识别[J].软件学报,2006,17(3):525-534.
[75]罗亚兰,陈锻生.基于彩色Gabor特征的人脸识别技术[J].中国图象图形学报,2008,13(2):242-246.
[76]孙雪梅,苏菲,蔡安妮.基于分叉树和SVM的人脸图像光照方向估计[J].中国图象图形学报,2007,12(10):1885-1888.
[77]何晓光,田捷,毋立芳,张瑶瑶,杨鑫.基于形态学商图像的光照归一化算法[J].软件学报,2007,18(9):2318-2325.
[78]张宝昌,陈熙霖,山世光,高文.基于支持向量的Kernel判别分析[J].计算机学报,2006,29(12):2143-2149.
[79]孔万增,朱善安.中心近邻嵌入学习算法的人脸识别研究[J].中国图象图形 学报,2008,13(4):691-695.
[80]杨欣,费树岷,陈丽娟.基于类矩阵和特征融合的加权自适应人脸识别[J].中国图象图形学报,2008,13(5):930-936.
[2]R.Chellappa,C.L.Wilson and S.Sirohey.Human and machine recognition of faces:A survey[J].IEEE Proc.1995,83(5):705-740.
[3]A.Samal and P.A.Iyengar.Automatic recognition and analysis of human faces and facial expressions:A survey[J].Pattern Recognition.1992,25(1):65-77.
[4]W.Zhao,R.Chellappa,P.J.Philips and A.Rosenfel.Face Recognition:A Literature Survey [J].ACM computring Surveys.2003,35(4):399-458.
[5]贾云得.机器视觉[M].1999科学出版社.
[6]Vapnik V N.An Overview of Statistical Learning Theory[J].IEEE Trans.On Neural Networks.1999,10(5):988-999.
[7]周激流,张哗.人脸识别理论研究进展[J].计算机辅助设计与图像学报.1999,11(2):23-29.
[8]Erik.Hjellnas,Boon.KeeLow.Face Detcetion:ASurvey[J].Computer Vision and Image understanding.2001,83:236-274.
[9]Y.Amit,D.Geman,B.Jedynak.Efficient focusing and face detection[J].Face Recognition.1998.
[10]梁路宏,艾海舟,徐光佑.人脸检测研究综述[J].计算机学报.2002.
[11]D.Demers,G.W.Cottrell.Nonlinear Dimensionality Reduetion[A].Advances in Neural Information Processing System[C],San Mateo,CA:Morgan Kaufinann.1993,580-587.
[12]Huang C L,Chen C W.Human facial feature extraction for face interpretation and Recognition[J].Pattern Recognition.1992,25(2):1435-1444.
[13]Kass M,Witkin A,Terzopoulos D.Snakes.Active contour models[J].International Journal of Computer Vision.1988,1(4):321-331.
[14]Sirovich L,Kirby M.Application of Karhunen-Loeve procedure for the characterization of human faces[J].IEEE Trans.PAMI,1990,3(1):71-79.
[15]Turk M A,Pentland A P.Eigenfaces for recognition[J].Journal of Cognitive Neuroscience,1991,3(1):71-86.
[16]Paul D.A neural network for facial feature location[J].UC Berkeley CS283 Project Report.1992,10:1-9.
[17]刘小军.人脸识别技术研究[D].中国科学院.2001.
[18]Lee S Y,Ham Y K,Park R H.Recognition of human front faces using knowledge-based feature extraction and neuro-fuzzy algorithm[J].Pattern Recognition,1996,29(11):1863-1879.
[19]Valentin D,Abdi H,O'Toole A,Cottrell G W.Connectionist models of face processing:A survey.Pattern Recognition.1994,27(9):1209-1230.
[20]Intrator N,Reisfeld D,Yeshurun Y.Face recognition using a hybrid supervised/unsupervised neural network[J].Pattern Recognition Letters.1996,17(1):67-76.
[21]Sam T,Roweis and Saul K L.Nonlinear dimensionality reduction by locally linear embedding[J].Science.2000,290(5500):2323-2326.
[22]He X and Niyogi.Locality preserving projections[C].Proceedings of Advances In Neural Information Processing Systems 16,MA:Cambridge,MIT Press.2004,153-160.
[23]Zhao Haitao,Sun Shaoyuan,Jing Zhongliang,and Jingyu Yang.Local structure based supervised feature extraction[J].Pattern Recognition.2006,39(881):1546-1550.
[24]M.Lades,J.C.Vorbruggen,J.Buhmann,J.Lange.Distortion invariant object recognition in the dynamic link architecture[J].Transcation Computer.1993,42:300-311.
[25]C.Liu and H.Wechsler.Gabor feature based classification using the enhanced fisher linear discriminant model for face recognition[J].Transaction on Image Processing.2002,11(4):467-476.
[26]Abhijit S.Pandya,Robert B.Macy著.徐勇,荆涛等译.神经网络模式识别及其实现[M].北京:电子工业出版社,1999:153-178.
[27]Gutta S,Huang J,Singh D.Benchmark.Studies on Face Recognition.Proceedings of International Workshop on Automatic Face and Gesture Recognition.IWAFGR 95,1995:227-231.
[28]S.Z.L,J.Lu.Face recognition using the nearest feature line method[J].Transactions on Neural Networks.11999,10(2):439-443.
[29]Burges C JC.A tutorial on support vector machines for pattern recognition[J].Data Mining and Knowledge Discovery.1998,2(2):57-66.
[30]SehoLkopf,Burges.VapnikV.Extracting Support Data for a Given Task[A]Proc.Of 1~(st) International Conference on Knowledge Discovery and Data Mining[C],AAAI,1995.252-257
[31]Vavnik V.The Nature of Statistical Learning Theory(Zndedition.)[M].New York,2000.
[32]Boser B,Guyonl,VapnikV.A Training Algorithm for Optimal Margin classifiers [A].Proceedings of Annual Workshop on Computational Learning Theory[C].Canada,1992.
[33]J.A.K.Suykens,J.Vandewalle.Last Squares Support Vector machine Classifiers[J].Neural processing Letters.1999,9(3):293-300.
[34]PeterAndras.The Equivalence of Support Machine and Regularization Neural Networks[J].Neural processing Letters.2002,15(2):97-104.
[35]刘江华,程君实,陈佳品.支持向量机训练算法综述.信息与控制.2002
[36]Chih WeiHsu.A companies on of methods for multi-class support vector machines[J].Trans Neural Networks.2002,13(2):415-425.
[37]U.KreBel.Pair wise classification and support vector machines[J].Advances in Kernel Methods:Support vector Leaning.1999,255-268.
[38]潘翔.基于多分类器集成的模式识别研究[D].2002,浙江工业大学.
[39]Killter J,Hatef M,Duin R P Wet al.On Combining Classifiers[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.1998,20(3):226-239.
[40]叶俊勇,汪同庆.基于支持向量机的人脸检测算法[J].计算机工程.2003,29(2):23-25.
[41]P.J.Phillips,H.Wechsler,J.Huang,P.Rauss,"The FERET database and evaluation procedure for face recognition algorithms," Image and Vision Computing J,Vol.16,No.5,pp.295-306,1998.
[42]P.J.Phillips,H.Moon,S.A.Rizvi,P.J.Rauss,"The FERET Evaluation Methodology for Face Recognition Algorithms," IEEE Trans.Pattern Analysis and Machine Intelligence,Vol.22,pp.1090-1104,2000.
[43]Dang-Hui Liu,Kin-Man Lain,Lan-Sun Shen.Optimal Sampling of Gabor Features for Face Recognition[J].Pattern Recognition Letters,2004,25:267-276.
[44]K.Jonsson,J.Matas,J.Kittler,Y.P.Li.Learning Support Vectors for Face Verification and Recontion[C].Proceeding of the 4th Int.Conf.on Auto.Face and GestureRecog:208-213,Grenoble,France,2000.3.
[45]Schapire,R.The Strength of Weak Leanability[J].Machine Leaning,1990,5(2):197-227.
[46]PengYang,Shiguang Shan,Wen Gao,Stan Z.Li,Dong Zhang.Face Recognition Using Ada-Boosted Gabor Features[J].Proc.Of Int.Conf.on Auto.Face and Gesture recognition,2004.
[47]Qijun Zhao,Hongtao Lu,and David Zhang.A Fast Evolutionary Pursuit Algorithm based on Linearly Combining Vectors[J].Pattern Recognition,2006,39:310-312.
[48]W.S.Zheng,J.H.Lal,and P.C.Yuen,GA-Fisher.A new LDA-based face recognition algorithm with selection of principal components[J].IEEE Transaction on Systems,Man,and Cybernetics-Part B,2005,5(35):1065-1078.
[49]D.Goldberg.Genetic Algorithms in Search,Optimization,and Machine Learning[M].Adison-Wesley,1989.
[50]R.Brunelli,T.Poggio.Face Recognition:Features vs.Templates[J].IEEE Trans.On PAMI,1993,15(10):1042-1052.
[51]B.Moghaddam.Principal Manifolds and Probabilistic subspaces for visual recognition[J].IEEE Trans.PAMI,2002,24(6) 780-788.
[52]B.Moghaddam and A.Pentland,Probabilistic Visual Learning for Object Representation[J].IEEE Trans.On PAMI,1997,20(7):696-710.
[53]L.Back,B.A.Draper,J.R.Beveridge and Kai She.PCA vs.ICA:A comparison on the FERET data set[C].Int'l Conf.on Computer Vision,Pattern Recognition and Image Processing,2002.
[54]P.Common.Independent Component Analysis-A New Concept for Signal Processing,1994,36:287-314.
[55]Yang,Miang-Hsuan,Kriegman,David J.,Ahuja,Narendra.Dectecting faces in images:A survey,Source[J].IEEE Tran.On PAMI,2002,24:34-58.
[56]Q.J.Zhao and H.T.Lu.GA-Driven LDA in KPCA Space for Facial Expression Recognition[J].Lecture Notes on Computer Science,2005,3611:8-36.
[57]Ming.Hsuan Yang.Recent Advances in Face Detection[R].IEEE.ICPR, 2004,Tutorial,Cambridge,UK.
[58]Chen,S.,Zhang,D.,and ZhouZ.-H.A new face recognition methodbased on SVD perturbation for single example image per person[C].Appl.Math.Comput,2005.
[59]徐科军,于翠欣.基于DSP的柯氏质量流量计信号处理系统[J].仪器仪表学报,2002,23(2):170-177.
[60]金忠,杨静宇,陆建峰.一种具有统计不相关的最佳鉴别矢量集[J].计算机学报,1999,22(10).
[61]边肇祺,张学工.模式识别(第二版)[M].北京:清华大学出版社,2000.
[62]梁路宏,等.基于多关联模板匹配的人脸检测[J].软件学报,2001,12(1):92-102.
[63]祝磊,朱善安.基于二维广义主成分分析的人脸识别[J].浙江大学学报,2007,41(2):264-267.
[64]陈伏兵,陈秀宏,高秀梅,等,二维主成分分析方法的推广及其在人脸识别中的应用[J].计算机应用,2005,25(8):1767-1770.
[65]王跃明.表情不变的三维人脸识别研究[D].浙江大学计算机学院博士学位论文,2007.
[66]刘向东.全自动三维人脸识别框架[D].吉林大学硕士学位论文,2007.
[67]赖剑煌,颜鑫弘,邓东皋.完善频谱脸人脸识别的分类器设计[J].中国图像图形学报,2002,7(5):466-471.
[68]赖剑煌,阮邦志,冯国灿.频谱脸.一种基于小波变换和Fourier变换的人脸识别新方法[J].中国图像图形学报,1999,4(10):811-817.
[69]杨琼,丁晓青.对称主分量分析及其在人脸识别中的应用[J].计算机学报,2003,26(9):1146-1151.
[70]刘青山,卢汉清,马颂德.综述人脸识别中的子空间方法[J].自动化学报,2003,29(6):900-911.
[71]彭岩,张道强.半监督典型相关分析算法[J].软件学报,2008,19(11):2822-2832.
[72]唐亮,熊蓉,褚健.单样本条件下权重模块2DPCA人脸识别[J].中国图象图形学报,2008,13(12):2307-2313.
[73]张莹,王耀南.基于Gabor滤波器包络的人脸识别算法[J].中国图象图形学报,2008,13(12):2314-2320.
[74]柴秀娟,山世光,卿来云,陈熙霖,高文.基于3D人脸重建的光照、姿态不变人脸识别[J].软件学报,2006,17(3):525-534.
[75]罗亚兰,陈锻生.基于彩色Gabor特征的人脸识别技术[J].中国图象图形学报,2008,13(2):242-246.
[76]孙雪梅,苏菲,蔡安妮.基于分叉树和SVM的人脸图像光照方向估计[J].中国图象图形学报,2007,12(10):1885-1888.
[77]何晓光,田捷,毋立芳,张瑶瑶,杨鑫.基于形态学商图像的光照归一化算法[J].软件学报,2007,18(9):2318-2325.
[78]张宝昌,陈熙霖,山世光,高文.基于支持向量的Kernel判别分析[J].计算机学报,2006,29(12):2143-2149.
[79]孔万增,朱善安.中心近邻嵌入学习算法的人脸识别研究[J].中国图象图形 学报,2008,13(4):691-695.
[80]杨欣,费树岷,陈丽娟.基于类矩阵和特征融合的加权自适应人脸识别[J].中国图象图形学报,2008,13(5):930-936.