摘要
人脸识别作为计算机视觉以及模式识别研究的一个重要子领域,它具有重要的理论研究意义和实际应用价值。近年来,人脸识别系统已经可以较为准确地在某些限定的条件下对人脸图像进行识别,但在实际应用中还面临着很多困难,其采用的识别方法尚需进一步改进和完善。
鉴于此,本文对人脸识别技术中的三种主流方法(PCA方法、贝叶斯方法和LDA方法)进行了深入的研究,提出了相应的改进方法。本文的主要贡献如下:
(1)PCA是重要的降维方法之一,它能用空间的主成分来逼近原空间。通常,PCA需要将一幅图像转化为一个高维向量,这个高维向量生成一个高维的协方差矩阵,这个高维的协方差矩阵导致计算的复杂性增加和存储空间也增大。和经典的PCA相比,2DPCA直接利用二维图像计算协方差矩阵,这个协方差矩阵的维数较小,它减少了计算的复杂性和节约了存储空间。尽管2DPCA促进了人脸识别技术的发展和进步,但相对于PCA生成的高维的协方差矩阵而言,2DPCA生成的协方差矩阵空间结构小,丢失了一些信息,这些丢失的信息对人脸是识别非常重要。为了利用更多的鉴别信息,本文提出了一种人脸垂直对称的变形2DPCA算法(S2DPCA),该算法与PCA相比降低了计算复杂性,与2DPCA和PCA相比提高了人脸识别率。
(2)因为S2DPCA比2DPCA有更高的自由度,所以它对小样本问题非常敏感。为了进一步提高S2DPCA方法的性能,本文提出了一种加权变形2DPCA的人脸特征提取方法(WV2DPCA)。该方法首先将人脸图像分为三个子图像(眉毛以上的部分、眉毛和鼻尖之间的部分和鼻尖以下的部分),接着对各子图像进行特征提取,最后根据每个子图像在人脸分类中的权重,利用加权最近邻分方法进行分类。该方法有三个优点:能减轻小样本的影响,提高人脸识别率,减小计算的复杂性。
(3) WV2DPCA方法中的权值大小是根据样本粗略估计,不够准确。为了解决以上问题,本文提出一种自适应加权变形2DPCA方法(AWV2DPCA),该方法直接将人脸图像分割成若干个子图像集(相同位置的子图像形成一个集合),然后根据相似性原理自适应地计算每个子图像集在分类中的权重,最后根据加权最近邻方法计算测试样本的类别。该方法的优点是:权重的大小自适应样本的类别。
(4)传统贝叶斯空间的人脸识别算法一般是假设样本空间满足高斯分布,实际上样本空间是很复杂的,并不一定满足高斯分布。为了适应人脸复杂的空间变化,本文提出了一种基于二值数据的贝叶斯子空间的人脸识别算法,该算法将图像二值化;然后假设各样本的特征空间变量相互独立,计算类条件概率;最后根据贝叶斯公式求后验概率。它克服了传统贝叶斯方法难求类内和类间协方差矩阵的缺点,简单易用。
为了进一步提高识别效果,提出了最小风险贝叶斯决策的二值化人脸识别算法,该算法根据图像的相似性估计其损失函数,利用贝叶斯公式求最小风险,最后根据最小风险判断其所属类。该算法增大类间距离,提高人脸正确识别率。
(5)线性判别分析法(LDA)在进行高维的人脸识别时,经常会出现“小样本问题(SSS)”和边缘类重叠问题。鉴于此,本文提出了一种可调控参数的LDA人脸识别方法。该方法重新定义了类内离散度矩阵,利用参数平衡其特征值估计的偏差和方差,从而解决小样本问题;对类间离散度矩阵加权,使边缘类均匀分布来防止边缘类的重叠,从而提高人脸正确识别率;实验表明,该方法可以解决小样本问题,且其性能优于传统的Eigenfaces和Fisherface等方法。在此基础上本文还提出了一种人脸本征空间的特征提取算法,该算法将类内离散度矩阵的特征空间分解为二个子解空间(主成分子空间和零子空间),利用本征谱模型对二个子空间进行正则化,从而减轻了不稳定性、过拟合和推广能力差的问题。实验表明,该算法使用较少的特征维数就能达到其它方法相同的识别率。
Face recognition, an important sub-field of computer vision and pattern recognition, has been extensively studied mainly due to its theoretical and practical significance. Although the face recognition has gained great progress, it includes still a lot of unsolved difficult problems. Hence, face recognition requires a further study for practical applications. This paper simply discusses the current research focus such as PCA, LDA and Bayesian methods, and proposes some improved methods for face recognition.
(1) PCA is one of the most important algorithms for dimension reduction, in which the original space is approximated by several principal components of all features so that mean square error (MSE) is minimized. Usually, PCA methods require that an image matrix is arranged into a high dimensional vector that will produce a very high dimensional covariance matrix which leads to two limits, extremely high computational complexity and very large storage space. Compared with the classical PCA,2DPCA directly uses two-dimensional images matrix to calculate a smaller (lower) covariance matrix, which decreases the computation load and saves the storage space. Although 2DPCA promotes the development of face recognition, some information contained in the high dimensional covariance matrix, which can help to improve face recognition rate, is lost by lower dimensional covariance matrix. To exploit more discriminant information, we propose a vertically symmetrical variation 2DPCA (S2DPCA) algorithm for face recognition. The experiments on face databases show that S2DPCA reduces the computational complexity comparing with PCA, and improves the face recognition rate comparing with PCA and 2DPCA.
(2) However, because S2DPCA has higher degree of freedom than 2DPCA, it is more sensitive to small sample size. To further improve the performance of the S2DPCA, an efficient algorithm of face feature extraction is proposed on basis of the weighted variation of 2DPCA (WV2DPCA), in which the face space is elaborately divided into three parts:the part above the eyebrows, one between eyebrows and the nasal tip, and that below nasal tip. WV2DPCA extracts the features from each sub-image matrix. According to taking the different roles of three sub-images in face recognition, a common weight is assigned to each sub-image. Finally, the classification is performed by the weighted nearest neighbor method. The WV2DPCA has three advantages that include alleviating the bad effect of small sample size, increasing face recognition rate, and decreasing the computational complexity.
(3) The main drawback in WV2DPCA is the strategy how to assign appropriate weights to the sub-images. However, weights used in WV2DPCA are roughly estimated according to different samples, and may be inaccurate. To overcome this drawback, an adaptive weighted variation 2DPCA (AWV2DPCA) for face recognition is developed. In AWV2DPCA, each face image is divided into several sub-images, and all the sub-images of the same position are defined as a sub-image set. AWV2DPCA extracts the features associated with each of sub-image sets, and adaptively estimates the weight corresponding to each of sub-image sets according to the similarity of features, and performs the classification by the weighted nearest neighbor method.
(4) The traditional Bayesian algorithms for face recognition assume that the samples meet the Gaussian distribution. In fact, the distributions of samples are very complicated. To adapt to the complicated distributions of samples, a face recognition algorithm based on the binary image is proposed in the light of the Bayesian principle. In this algorithm, the images binarization is firstly performed, and the class conditional probability is calculated under the assumption that the sample feature variables are mutually independent each other, and finally posterior probability is calculated by Bayesian formula. The above simple method decreases the computational complexity.
To further enhance the face recognition rate, a face recognition algorithm of binary image is proposed by the smallest Bayesian risk method, which estimates the loss function according to the similarity of the samples, then evaluates the smallest risk by Bayesian formula, and finally determines that they belong to which class. This algorithm increases the gaps between classes and improves face recognition rate.
(5) It is well-known that the line discriminant analysis (LDA) applied to high-dimensional face recognition often suffers from two problems, small sample size (SSS) and close-to-class overlap. To overcome these problems, a LDA-based face recognition algorithm with regularization parameters is proposed, which resolves the SSS problem by regularization parameters and redefined within-class scatter matrices, and prevents from the overlap of edge classes through weighting between-class scatter matrices. Extensive experimental results show the proposed LDA algorithm can solve the above two problems and outperforms traditional methods such as eigenfaces and Fisherfaces by controlling the regularization parameters. On the basis of the above method, an algorithm of feature extraction in face image space is proposed. This algorithm decomposes the eigenfeature space into two subspaces, principal component subspace and null subspace, and regularizes two subspaces with the eigen-feature spectrum model to alleviate the instability, overfitting or poor generalization. The experiments on ORL face database show that our method, which uses fewer features, can achieve higher recognition rate than other approaches, such as FLDA, BML, and DSL.
引文
[1]D. Maltoni, D. Maio, A. K. Jain, et al., Handbook of fingerprint recognition. Springer-Verlag,2003.
[2]A. Jain, L. Hong, R. Bolle, On-line fingerprint verification. IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19(4):302-314.
[3]Feng Guiyu, Hu Dewen, D. Zhang, et al., An alternative formulation of kernel LPP with application to image recognition. Neurocomputing,2006,69(13-15): 1733-1738.
[4]Jing Xiaoyuan, D. Zhang, A face and palmprint recognition approach based on discriminant DCT feature extraction. IEEE Transactions on systems, Man and Cybernetics,2004,34(6):2405-2415.
[5]D. Zhang, K. Wai Kin, J. You, et al. Online palmprint identification. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(9): 1041-1050.
[6]E. Hjelmas, B. K. Low, Face detection:A survey. Computer Vision and Image Understanding,2001,83(3):236-274.
[7]W. Zhao, R. Chellappa, P. J. Phillips, et al., Face recognition:A literature survey. Acm Computing Surveys,2003,35(4):399-459.
[8]R. P. Wildes, his recognition:An emerging biometric technology. Proceeding of the IEEE,1997,85(9):1348-1363.
[9]J. Daugman, How iris recognition works. IEEE Transactions on Circuits and System for Video Technology,2004,14(1):21-30.
[10]Z. M. Faungez, Signature recognition state-of-the-art. IEEE aerospace and electronic system magazine,2005,20(7):28-32.
[11]V. S. Nalwa, Automatic on-line signature verification. IEEE/IAFE/INFORMS Conference on computational intelligence for Financial Engineering,1997,85: 215-239.
[12]S. Mozaffari, K. Faez, F. Faradji, One dimensional fractal coder for online signature recognition. International conference on pattern recognition,2006, 2:857-860.
[13]J. Mantyjarvi, J. Koivumaki, P. Vuori, Keystroke recognition for virtual keyboard. IEEE international conference on multimedia and expo,2002,2: 429-432.
[14]International biometric group, Biometrics market and industry report 2007-1012. 2008.
[15]M. Turk, A. Pentland, Eigenfaces for recognition. Journal of cognitive neurOsc-ience,1991,3(1):71-86.
[16]M. Turk, A. Penland, Face recognition using eigenfaces. Proceedings of IEEE computer vision and pattern recognition, Hawaii, USA CS Press,1991,586-591.
[17]H. Peng, D. Zhang, Dual eigenspace method for human face recognition. IEEE transaction on electronics letters,1997,33(4):283-284.
[18]D. L. Sweet, J. J. Weng, Using discriminant eigenfeatures for image retrieval. IEEE transaction on pattern analysis and machine intelligence,1996,18(8): 831-836.
[19]K. Fukunaga, Introduction to Statistical Pattern Recognition. Academic Press, New York,1990.
[20]A. Levy and M. Lindenbaum, Sequential Karhunen-Loeve basis extraction and its application to images. IEEE Trans. on Image Processing,2000,9:1371-1374.
[21]Lei Xu, A. L. Yuille, Robust principal component analysis by self-organizing rules based on statistical physics approach. IEEE Transactions on Neural Networks,1995,6(1):131-143.
[22]Fernando De la Torre, Michael J. Black, Robust Principal Component Analysis for Computer Vision. Int. Conf. on Computer Vision (ICCV'2001), vol.1, Page(s):362-369, Vancouver, Canada, July 2001.
[23]D. Wang, J.A. Romagnoli, Robust multi-scale principal components analysis with applications to process monitoring. Journal of Process Control,2005,15(8): 869-882.
[24]I. Stanimirova, B. Walczak, D. L. Massart and V. Simeonov. A comparison between two robust PCA algorithms. Chemometrics and Intelligent Laboratory Systems,2004,71(1):83-95.
[25]A. Mike Burton, Rob Jenkins, Peter J. B. Hancock and et al., Robust representations for face recognition:The power of averages. Cognitive Psychology,2005,51(3):256-284.
[26]Zhao Shijian, Xu Yongmao, Multivariate Statistical Process Monitoring Using Robust Nonlinear Principal Component Analysis. Tsinghua Science & Technology,2005,10(5):582-586.
[27]Xu Yong, Z. David, Song Fengxi and et al., A method for speeding up feature extraction based on KPCA. Neurocomputing,2007,70(4-6):1056-1061.
[28]K. Das, S. Osechinskiy, Z. Nenadic, A classwise PCA-based recognition of neural data for brain-computer interfaces. In Proceedings of the 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society,2007:6519-6522.
[29]K. Das, Z. Nenadic, An efficient discriminant-based solution for small sample size problem. Pattern Recognition,2009,42:857-866.
[30]M. Soo Park, J. Young Choi, Theoretical analysis on feature extraction capabi-lity of class-augmented PCA. PatternRecognition,2009,42:2353-2362.
[31]Yang Jian, Yang Jingyu, From image vector to matrix:A straight forward image projection technique-IMPCA vs. PCA. Pattern Recognition,2002,35(9): 1997-1999.
[32]M. Benito and D. Pena, A fast approach for dimensionality reduction with image data. Pattern Recognition,2005,38(12):2400-2408.
[33]P. Nagabhushan, D.S. Guru and B.H. Shekar, Visual learning and recognition of 3D objects using two-dimensional principal component analysis:A robust and an efficient approach. Pattern Recognition,2006,39(4):721-725.
[34]Zuo Wangmeng, Wang Kuanquan and Z. David, An assembled matrix distance metric for 2DPCA-based image recognition. Pattern Recognition Letters,2006, 27(3):210-216.
[35]Chen Songcan, Zhu Yulian., Zhang Daoqiang and et al., Feature extraction approaches based on matrix pattern:MatPCA and MatFLDA. Pattern Recognition Letters,2005,26(8):1157-1167.
[36]Wang Liwei, Wang Xiao, Feng Jufu, On image matrix based feature extraction algorithms. IEEE Transactions on Systems, Man and Cybernetics, Part B,2006, 36(1):194-197.
[37]R. P. W. Duin, M. Loog, T. K. Ho. Recent submissions linear dimensionality reduction and face recognition. Pattern Recognition Letter,2006,27(7): 707-708.
[38]Yan Shuicheng, Xu Dong, Zhang Benyu, and et al., Graph embedding and extensions:a general framework for dimensionality reduction. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29:40-51.
[39]Yang Jian, Liu Chengjun, Horizontal and Vertical 2DPCA-based Discriminant analysis for Face Verification on a Large-scale Database, IEEE Transactions on Information Forensics and Security,2007,2(4):781-792.
[40]He Xiaofei, Cai Deng, N. Partha, Tensor Subspace Analysis. Advances in Neural Information Processing Systems,2005.
[41]Wang Hongcheng, A. Narendra, A tensor approximation approach to dimensiona-lity reduction. Journal of Computer Vision,2008,76:217-229.
[42]P.N. Belhumeur, J.P. Hespanha, D.J. Kriegman, Eigenfaces vs fisherfaces: Recognition using class specific linear projection. IEEE Transactions on pattern analysis and machine intelligence,1997,19(7):711-720
[43]S.S. Wilks, Mathematical Statistics, Wiley, New York,1962.
[44]D. Richard O., H. Peter E., and S. David G., Pattern Classification, John Wiley & Sons, seconded.,2001.
[45]Zhong Jin, Yang Jingyu, Hu Zhongshan, and et al., Face Recognition based on uncorrelated discriminant transformation. Pattern Recognition,2001,34(7): 1405-1416.
[46]Jin Zhong, Yang Jingyu, Tang Zhenmin, and et al., A theorem on the uncorrelated optimal discriminant vectors. Pattern Recognition,2001,34(10):2041-2047.
[47]R.P.W. Duin, M. Loog, Linear dimensionality reduction via a heteroscedastic extension of LDA:the Chernoff criterion. IEEE Trans. PAMI,2004,26(6): 732-739.
[48]D. D. Ridder, M. Loog, M.J.T. Reinders, Local Fisher embedding, Proc.17th International Conference on Pattern Recognition (ICPR2004),2004,295-298.
[49]Li Haifeng, Jiang Tao, Zhang Keshu, Efficient and robust feature extraction by maximum margin criterion. In Proc. of Neural Information Processing Systems, 2003.
[50]Song Fengxi, Yang Jingyu, Liu Shuhai, Large margin linear projection and face recognition. Pattern Recognition,2004,37(9):1953-1955.
[51]Song Fengxi, D. Zhang, Mei Dayong, and et al., A multiple maximum scatter difference discriminant criterion for facial feature extraction. IEEE Transactions on SMC, Part B,2007,37(6):1599-1606.
[52]K. Keunchang, P. Witold, Face recognition using a fuzzy Fisher face classifier. Pattern Recognition,2005,38(10):1717-1732.
[53]Zhuang Xiaosheng, Dai Daoqing, Inverse Fisher discriminant criteria for small sample size problem and its application to face recognition. PatternRecognition, 2005,38(11):2192-2194.
[54]Yang Wankou, Wang Jianguo, Ren Mingwu and et al., Feature extraction using fuzzy inverse FDA. Neurocomputing,2009,72(13-15):3384-3390.
[55]J.H. Friedman, Regularized discriminant analysis. J. Am. Stat. Assoc.,1989,84 (405):165-175.
[56]Hong Ziquan, Yang Jingyu, Optimal discriminant plane for a small number of samples and design method of classifier on the plane. Pattern Recognition,1991, 24(4):317-324.
[57]Cheng Yongqing, Zhuang Yongming, Yang Jingyu, Optimal Fisher discriminant analysis using the rank decomposition. Pattern Recognition,1992,25(1): 101-111.
[58]T. Hastie, R.Tibshirani, Penalized discriminant analysis.The Annals of Statistics, 1995,23(1):73-102.
[59]P. N. Belhumeur, Eigenfaces vs. Fisherfaces:recognition using class speci9c linear projection. IEEE Trans. Pattern Anal.Mach. Intell.,1997,19(7):711-720.
[60]Chen Lifen, Mark Liao Hongyuan, Ko Mingtat, and et al., A new LDA-based face recognition system which can solve the small sample size problem. Pattern Recognition,2000,33:1713-1726.
[61]Yu Hua, Yang Jie, A direct LDA algorithm for high-dimensional data with application to face recognition. Pattern Recognition,2001,34(11):2067-2070.
[62]Yang Jian, Yang Jingyu, Why can LDA be performed in PCA transformed space?. Pattern Recognition,2003,36:563-566.
[63]P. Howland, H. Park. Generalizing discriminant analysis using the generalized singular value decomposition, IEEE Trans. Pattern Anal. Mach. Intell.,2004,26 (8):995-1006.
[64]C.E. Thomaz, D.F. Gillies, R.Q. Feitosa, A new covariance estimate for Bayesian classifier in biometric recognition. IEEE CSVT,2004,14(2):214-223.
[65]Liu Chaochun, Dai Daoqing, Yan Hong, Local Discriminant Wavelet Packet Coordinates for Face Recognition. Journal of Machine Learning Research,2007, 8:1165-1195.
[66]M. Kyperountas, A. Tefas, I. Pitas, Weighted Piecewise LDA for Solving the Small Sample Size Problem in Face Verification. IEEE Transactions on Neural Networks,2007,18(2):506-519.
[67]M. Sebastian, R. Gunnar, M. Klaus Robert, A mathematical programming approach to the Kernel Fisher algorithm. In T.K. Leen, T.G. Dietterich, and V. Tresp, editors, Advances in Neural Information Processing Systems 13, MIT Press,2001:591-597.
[68]G. Baudat, F. Anouar, Generalized discriminant analysis using a kernel approach. Neural Computation,2000,12 (10):2385-2404.
[69]Y. Koren, L. Carmel, Robust linear dimensionality reduction. IEEE Trans. On Visualization and Computer Graphics,2004,10(4):459-470.
[70]Yang Jian, Zhong Jin, Yang Jingyu, and et al, The essence of kernel Fisher discriminant:KPCA plus LDA. Pattern Recognition,2004,37(10):2097-2100.
[71]S. Zafeirio, A. Tefas, I. Pitas, Minimum Class Variance Support Vector Machines. IEEE Transactions On Image Processing,2007,16(10): 2551-2564.
[72]Yang Jian, Z. David, Xu Yong and et al, Two-dimensional discriminant transform for face recognition. Pattern Recognition,2005,38(7):1125-1129.
[73]Vo Dinh Minh Nhat, Sung Young Lee, Two-dimensional weighted PC A algorithm for face recognition. Proceedings of the IEEE International Symposium on Computational Intelligence in Robotics and Automation (CIRA 2005),2005:219-223.
[74]Jing Xiaoyuan, Wong Hausan, Z. David. Face recognition based on 2D Fisherface approach. Pattern Recognition,2006,39(4):707-710.
[75]Qiu Xipeng, Wu Lide, Two-dimensional nearest neighbor discriminant analysis. Neurocomputing,2007,70(13-15):2572-2575.
[76]Zhi Ruicong, Ruan Qiuqi, Two-dimensional direct and weighted linear discri-minant analysis for face recognition. Neurocomputing, Volume 71, Issues 16-18, October 2008, Pages 3607-3611.
[77]Liang Zhizheng, Li Youfu, Shi Pengfei, A note on two-dimensional linear discriminant analysis. Pattern Recognition Letters,2008,29(16):2122-2128.
[78]Zhi Weishi, J. H. Lai, Stan Z. Li,1D-LDA vs.2D-LDA:When is vector-based linear discriminant analysis better than matrix-based?. Pattern Recognition, 2008,41(7):2156-2172.
[79]Wang Jianguo, Yankou Wankou, Lin Yusheng, and et al, Two-directional maximum scatter difference discriminant analysis for face recognition. Neurocomputing,2008,72(1-3):352-358.
[80]Yang Wankou, Wang Jianguo, Ren Mingwu, and et al., Feature extraction based on Laplacian bidirectional maximum margin criterion. Pattern Recognition, 2009,42(11):2327-2334.
[81]Yan Shuicheng, Xu Dong, Yang Qiang, and et al., Discriminant Analysis with Tensor Representation. In Proceedings of CVPR,2005.
[82]Zhang Wei, Lin Zhouchen, Tang Xiaoou. Tensor linear Laplacian discrimination (TLLD) for feature extraction. Pattern Recognition,2009,42:1941-1948.
[83]Nie Feiping, Xiang Shiming, Song Yangqiu, and et al., Extracting the optimal dimensionality for local tensor discriminant analysis. Pattern Recognition,2009, 42:105-114.
[84]S. Bernhard, S. Alexander, M. Klaus Robert, Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation,1998,10(5):1922-1319.
[85]Yang Ming Huan, Kernel Eigenfaces vs. kernel Fisherfaces:face recognition using kernel methods. In:Proceedings of international conference on automatic face and gesture recognition, USA, Washington DC.2002:215-220.
[86]B. Heisele, H. Purdy, T. Poggio, Face recognition with support vector Machines: Global Versus Component-based Approach, International Conference on Computer Vision (ICCV'01), Vancouver Canada,2001 (2):688-694.
[87]G. Guodong, L. Stan Z., C. Kapluk, Face recognition by Support Vector machines. Proc. Of the 4th Int. Conf. on Auto, Face and Gesture Recog.2000: 196-201.
[88]O. Deniz, M. Castrillon, M. Hemandez, Face recognition using independent component analysis and support vector machines. Pattern recognition letters, 2003,24:2153-2157.
[89]Wang Yingjie, C. Chin Seng, H. Yeong Khing., Facial feature detection and face recognition from 2D and 3D images. Pattern Recognition letters,2002,23: 1191-1202.
[90]R. Chellappa, C. Wilson, S. Sirohet, Hum and machine recognition of faces a survey. Proc of the IEEE,1995,83(5):705-740.
[91]R. Brunellir, T. Poggid, Face recognition features versus template. IEEE Transactions on pattern analysis & machine intelligence,1993,15(10): 1042-1052.
[92]邵平,杨路明,黄海滨等,基于积分图像的快速模板匹配.计算机科学,2006,33(12):225-229.
[93]Zhang Jun, Yan Yong, M. lades, Face recognition eigenface, elastic matching and neural nets. Proc. of the IEEE,1997,85(9):1422-1435.
[94]L. Wiskott, J.M. Fellous, N. Kruger, and, et al., Face recognition by elastic bunch graph matching. IEEE Transactions on pattern analysis & machine intelligence,1997,19(7):775-779.
[95]A. Lanitis, C.J. Taylor, T.F. Cootes, Automatic interpretation and coding offace images using flexible models. IEEE Transactions on pattern analysis & machine intelligence,1997,19(7):743-756.
[96]B. Moghaddam, T. Jebara, A. Pentland, Bayesian face recognitiontion. Pattern recognition,2000,33(11):1771-1782.
[97]P. Penio S, A. Joseph J, Local feature analysis:a general statistical theory for object representation. Network computation in neural systems,1996,7(3): 477-500.
[98]Liu Chengjun, H. Wechsler, A gabor feature classifier for face recognition. Proc of the 8th international conference computer vision,2001:270-275.
[99]Wang Xiaogang, Tang Xiaoou. Bayesian face recognition based on Gaussian mixture models. Proc of the 17th international conferenceon pattern recognition, 2004:142-145.
[100]P. J. Phillips, M. Hyeonjoonn, S.A. Rizvi, and et al., The FERET evaluation methodology for face recognition algorithms. IEEE Transactions on pattern analysis & machine intelligence,2000,22(10):1090-1104.
[101]L. R. Rabiner, A tutorial on hidden markov models and selected application inspeech recognition. Proceeding of the IEEE,1989,77(2):257-285.
[102]V.N. Ara, H.H. Monson, A hidden Markov model-based approach for face detection and recognition. Georgia:Institute of technology,1999,38-108.
[103]A.V. Nefian, M.H. Hayes, An embedded HMM-based approach for face detection and recognition. IEEE international conference on acoustic speech and signal process, Phoenix, USA:IEEE CS Press,1999,2553-3556.
[104]H. Othman, T. Aboulnasr, A separable low complexity 2D HMM with application to face recognition. IEEE Transactions on pattern analysis and machine intelligence,2003,25(10):1229-1238.
[105]T. Kohonen, Self-organization and associative memory. Springer-verlag,1988.
[106]G.W. Cotterll, M.K. Fleming, Face recognition using unsupervised feature extraction. International Neural Network Conference,1990,322-325.
[107]S. Lawrence, C.L. Giles, A. C. tsol, and et al. Face recognition:A convolution-al neural-network approach. IEEE Transaction on neural network,1997,8(1): 98-113.
[108]Liu Shanghung, Kung Sunyuan, Liu Longji, Face recognition/detection by probabilistic decision-based neural network. IEEE Transactions on neural networks,1997,8(1):114-132.
[109]J. E. Meng, Wu Shiqian, Lu Juwei, and et al., Face recognition with radial basis function (RBF) neural networks. IEEE Transaction on neural networks, 2002,13(3):697-710.
[110]M. Smith, Neural networks for statistical modeling. International Thomson Computer press,1996.
[111]S. G. Tzafestas, P. J. Dalianais, G. Anthopoulos, and et al. On the overtraining phenomenon of backpropagation neural networks. Mathematics and computers in simulation,1996,40(5-6):507-521.
[112]M. Nishant, V. Benjamin, Principal-component-analysis-based estimation of blood flow velicities using optical coherence tomography intensity signals. Optics letters,2011,36(11):2068-2070.
[113]Yang jian, Z. David, F. F. Alejandro, and et al, Two-dimensional PCA:A new approach to appearance-based face representation and recognition. Journal of IEEE transaction on pattern analysis and machine intelligence,2004,26(1): 131-137.
[114]V. Muriel, C. Christophe, L. Christophe, Comparing robustness of two dimens-ional PCA and eigenfaces recognition. Lecture notes comput. Sci.,2004, 32(12):717-724.
[115]Z. Xianwu, G. lei, Face recognition based on subpattern two directional 2DPCA. Journal of optoelectronics laser,2009,20(11):1498-1502.
[116]Zhang Daoqiang, Zhou Zhihuan, (2D)2PCA:Two-directional two-dimensional PCA for efficient face representation and recognition. Neurocomputing,2005, 69:224-231.
[117]Zhang Longxiang, Face recognition method using improved modular 2DPCA. Computer engineering and application,2010,46(13):147-150.
[118]Yuan Ning, Wu Xiaojun, Wang Shitong, and et al., A face verification algorithm based on combination of modular 2DPCA and CSLDA. Jounal of computer research and development,2008,45(6):1029-1035.
[119]Kong Hui, Wang Lei, T. Earn Khwang, and et al., Generalized 2D principal component analysis for face image representation and recognition. Neural Networks,2005,18:585-594.
[120]Zeng Yue, Feng Dazheng, Xing Li, An Algorithm of Face Recognition Based on the Variation of 2DPCA. Journal of Computational Information Systems. 2011,7(1):303-310.
[121]Li Zhao, Yang Yeehong, Theoretical analysis of illumination in PCA-based vision system. Pattern recognition,1999,32(4):547-564.
[122]Y. Choi, T. Tokumoto, M. Lee, and et al, Incremental two-dimensional two-directional principal component analysis for face recognition. IEEE international conference on acoustics, speech and signal processing,2011, 1493-1496.
[123]G Rajkiran, K. A. Vijayan, An improved face recognition technique based on modular PCA approach. Pattern recognition letters,2004,25:429-436.
[124]Z. Wangmeng, D. Zhang, W. Kuanquan, Bidirectional PCA with assembled matrix distance metric for image recognition. IEEE Transaction on Systems, Man, and Cybernetics—Part B:Cyebrnetics,2006,36(4):863-872.
[125]C. Songcan, Z. Yulian, Subpattern-based principle component analysis. Pattern Recognition,2004,37(5):1081-1083.
[126]杨万扣,吉善兵,任明武等,基于增强的2维主成分分析的特征提取方法及其在人脸识别中的应用.计算机图像图形学报,2009,14(2):227-232.
[127]Zhao Dongjuan, Liang Jiuzhen, Face recognition algorithm fusing 2DPCA and fuzzy 2DLDA. Journal of computer application,2011,31(2):420-422.
[128]Huang Guohong, Fusion (2D)2PCALDA:A new Method for Face Recognition. Applied Mathematics and Computation,2010,216:3195-3199.
[129]Qi Yongfeng, Zhang Jiashu, (2D)2PCALDA:An Efficient Approach for Face Recognition. Applied Mathematics and Computation,2009,213:1-7.
[130]W. Jin, B. Armando, W. Lu, and et al., Multilinear principal component analysis for face recognition with fewer features. Neurocomputing,2010,73: 1550-1555.
[131]Yu Wangxin, Wang Zhizhong, Chen Weiting, A new framework to combine vertical and horizontal information for face recognition. Neurocomputing,2009, 72:1084-1091.
[132]C.Y. Pong, J.H. Lai, Face Representation Using Independent Component Analysis. Pattern Recognition,2002,35(6):1247-1257.
[133]H. Gunther, The principal components of natural images revisited. IEEE transactionas on pattern analysis and machine intelligence,2006,28(5):673-683.
[134]V. Muriel, G. Chrishophe, L. Christophe, Comparing robustness of two-dimensional PCA and eigenfaces for face recognition. Lecture notes in computer science,2004,3212:717-724.
[135]L. Chong, L. Wangquan, L. Xiaodong, and et al., Double sides 2DPCA for face recognition. Lectures notes in computer science,2008,5226:446-459.
[136]W. Haixian, Structural two-dimensional principal component analysis for image recognition. Machine vision and applications,2009,22(2):433-438.
[137]K. Chunghoon, C. Chongho, Image covariance-based subspace method for face recognition. Pattern recognition,2007,40(5):1592-1604.
[138]N. Nam, Liu Wangquan, V. Svetha, Random subspace two-dimensional PC A for face recognition. Lecture notes in computer science,2007,4810:655-664.
[139]Pan Xin, Ruan Qiuqi, Palmprint recognition using Gabor feature-based (2D)2PCA. Neurocomputing,2008,71(13-15):3032-3036.
[140]Meng Jicheng, Zhang Wenbin, Volume measure in 2DPCA-based face recognition. Pattern recognition letter,2007,28:1203-1208.
[141]Y. Hongchuan, M. Bennamoun,1D-PCA,2D-PCA to nD-PCA. The 18th international conference in pattern recognition,2006.
[142]L. Haiping, K. N. Plataniotis, A. N. Venetsanopoulos, MPCA:multilinear pricipal component analysis of tensor objects. IEEE transactions in neural networks,2008,19(1):18-39.
[143]李晓东,费树岷.一种改进的模块化2DPCA人脸识别方法.系统仿真学报,2009,21(15):4672-4675.
[144]曾岳,冯大政.一种基于加权变形的2DPCA的人脸特征提取方法.电子与信息学报,2011,33(4):763-768.
[145]Zhao Haitao, Yuen Pongchi, J.T. Kwok, A Novel Incremental Principal Component Analysis and Its Application for Face Recognition. IEEE Transactions on systems, man, and cybernetics—Part B:cybernetics.2006, 36(4):873-886.
[146]K. Young Gil, S. Young Jun, C. Un Dong, and et al., Face recognition using a fusion method based on bidirectional 2DPCA. Applied Mathematics and Computation.2008,205:601-607.
[147]Sun Yanfeng, Tang Hengliang, Yin Baocai, The 3D recognition algorithm fusing multi-geometry features. Acta automatica sinica,2008,34(12): 1483-1489.
[148]谢毓湘,王卫威,栾悉道等.基于肤色与模板匹配的人脸识别.计算机工程与科学,2008,30(6):54-59.
[149]蒋加伏,袁承伟.融合PCA与LDA变换的仿生人脸识别研究.计算机工程与应用,2010,46(19):160-163.
[150]赵武锋,沈海斌,严晓浪.直接LDA在人脸识别中的鉴别力分析.浙江大学学报(工学版),2010,44(18):1479-1483。
[151]顾明.基于模糊ART神经网络的在线人脸识别模型的设计和实现.计算机 科学,2007,34(8):232-235.
[152]武京伟,黄春庆.一种基于改进弹性束图匹配的人脸识别.工业控制计算机,2009,22(9):44-46.
[153]李健,李鹏坤, 师永刚.基于自由形状变形的三维人脸表情控制.计算机工程与科学,2010,32(3):59-62.
[154]Li Zhifeng, Tang Xiaoou, Using support vector machines to enhance the performance of bayesian face recognition. IEEE Transaction on information forensics and security.2007,2(2):174-180.
[155]张忠波,马驷良,董险峰.基于局部特征分析与最优化匹配的人脸识别算法.吉林大学学报(理学版),2005,43(1):59-63.
[156]赵韩,姜康,曹文钢等.用小波变换和Fisher判别对人脸进行特征提取.哈尔滨工业大学学报,2009,41(11):278-280.
[157]Wang Xiaogang, Tang Xiaoou, Bayesian face recognition based on Gaussian mixture models, Proc. of the 17 th International Conference on Pattern Recognition,2004:142-145.
[158]L. Juwei Lu, N. P. Kostantinos, N. V. Anastasios, Face recognition using LDA-based Algorithms. IEEE Trans on Neural networks,2003,14(1),117-126.
[159]E. Armin, F. Mohamad, H. A. Moghaddm, and et al., Block-wise 2D kernel PCA/LDA for face recognition. Information processing letters,2010,110: 761-766.
[160]T.V. Bandos, L. Bruzzone, G. Camps Valls, Classification of hyperspectral images with regularized linear discriminant analysis. IEEE transactions on geoscience and remote sensing,2009,47(3):862-873.
[161]E. K. Tang, P. N. Suganthan, X. Yao, et al., Linear dimensionality reduction using relevance weighted LDA. Pattern Recognition,2005,38:485-493.
[162]X. Jing Hao, D. M. Titterington, Do unbalanced data have a negative effect on LDA?. Pattern Recognition,2008,41:1558-1571.
[163]Yu Hua, Yang Jie, A direct LDA algorithm for high-dimensional data with application to face recognition. J. of Pattern Recognition,2001,34(10): 2067-2070.
[164]赵峰,张军英等.一种核Fisher判别分析的快速算法.电子与信息学报,2007,29(7):1731-1734.
[165]L. Juwei, N. P. Kostantinos, N. V. Anastasios, Regularized discriminate analysis for the small sample size problem in face recognition. Pattern recognition Letter,2003,24(16),3079-3087.
[166]M. Aleix M., Z. Manli, Where are linear feature extraction methods application?. IEEE trans. On Pattern analysis and machine intelligence,2005,27(1): 1934.1944.
[167]程正东,各毓晋,樊祥FISHER线性鉴别函数的一种推广形式.模式识别与人工智能,2009,22(2):176-181.
[168]Yang Jian, Yang Jingyu, Why can LDA be performed in PCA transformed space?. Pattern Recognition,2003,36(2):563-566.
[169]M. Kirby and L. Sirovich, Application of karhunen-loeve procedure for the characterization of human face. IEEE Trans on pattern analysis and machine intelligence,1990,12(1):103-108.
[170]B. Moghaddam, principal manifolds and probabilistic subspace for visual recognition. IEEE Trans on pattern analysis and machine intelligence,2002, 24(6):780-788.
[171]Liu Chengjun, H. Wechsler, enhanced fisher linear discriminate models for face recognition. Proceedings of IEEE international workshop on pattern recognition. Madison, Wisconsin, August 1998:1368-1372.
[172]程正东,各毓晋,樊祥.FISHER线性鉴别函数的一种推广形式.模式识别与人工智能,2009,22(2):176-181.
[173]J. L. Juwei, K. N. Plataniotis, A. N. Venetsanopoulos. Face recognition using LDA-based algorithms. IEEE Trans on neural networks,2003,14(1):195-200.
[174]Wang Xiaogang, Tang Xiaoou. Dual-Space linear discriminate analysis for face recognition. Proceedings of IEEE international workshop on computer vision and pattern recognition 2004,2:564-569.
[175]Liu Chengjun, H. Wechsler, Robust coding schemes for indexing and retrieval from larger face database. IEEE Trans on image processing.2000,9(1): 132-137.