可变光照下的人脸识别算法研究
|
摘要
人脸识别是图像处理和模式识别领域的一个重要研究课题，人脸识别和认证技术在公共安全、智能监控、多媒体等领域有着广阔的应用前景。经过数十年的研究，在理想情况下人脸识别技术已能取得较好的识别性能。但在不可控环境中易受到光照、姿态、表情、遮挡等因素的影响，使识别性能急剧下降。让人脸识别系统走向实用仍然是一个极具挑战性的课题。
本文针对光照问题进行研究，以提高人脸识别系统对光照的鲁棒性和识别率为主要目标，主要对人脸图像预处理、提取光照不变特征、分类识别等关键阶段展开研究，探讨人脸识别问题的研究方法。本文的主要工作和研究成果概括如下：
1.首先介绍了人脸识别课题的研究背景和意义及国内外发展情况，针对光照问题的三种研究方法：光照建模、光照补偿、光照不变特征提取法进行分类总结，分析其优缺点。
2.提出一种基于非下采样Contourlet变换（Nonsubsampled Contourlet Transform，NSCT）和邻域去噪的光照不变人脸识别算法。光照问题是影响识别效率的重要因素之一。本文在分析了小波变换提取人脸特征的基础上，采用多尺度的NSCT，它不仅具有小波变换的多分辨率和时频局域特性，还具有很强的方向性和冗余性，可以更完备的提取光照不变特征，同时在图像表示上能更好的描述人脸细节信息。采用邻域去噪方法去除光照不变特征中的投射阴影等噪声，因为投射阴影出现在局部的可能性最大，邻域去噪符合这点，其只在小范围内进行去噪，相比于一般的去噪方法，能保留更多的边界信息。经实验证明，该方法有效的提取光照不变特征，显著提高了人脸识别率。
3.人脸识别系统分为预处理、特征提取、分类识别三个关键环节，每一阶段对人脸识别系统性能都有所提升。本文提出的基于局部二值模式（Local Binary Pattern，LBP）和线性回归的可变光照人脸识别算法同时涉及这三个阶段，在每一阶段中对光照进行处理，得到更完备的可变光照算法。在预处理阶段采用Gamma校正、DoG滤波、对比度均衡化方法，降低光照敏感度，采用具有光照鲁棒性的分块LBP提取光照不变特征，最后使用改进的线性回归模型进行分类识别，即消除受光照影响最大的主成分系数。所提方法能有效的消除光照对人脸识别的影响，提高人脸识别系统的鲁棒性和识别率。
Face recognition is an important research topic in image processing and computervision, which has promising applications in public security, smart surveillance, multimediaand so on. Through face recognition has achieved great progress in the decades, when itcomes to uncontrolled conditions, such as different illumination conditions, pose variation,mixture of emotions and object shelter, and the accuracy of face recognition willdramatically decline. Therefore, how to build a real-life face recognition system is a sharpchallenging topic.
This paper focuses on the problem of face recognition in variant illuminationenvironment. The main purpose is to improve the recognition accuracy and robustness offace recognition system under various lighting condition. To achieve this purpose, thefocus of our work is on the image preprocessing, illumination invariant feature extraction,classification and recognition. The major work and contributions of this paper are follows:
1. The research background and significance of face recognition both at home andabroad are introduced firstly in this paper. Then the methods dealing with lighting problemare summarized, that is illumination model, preprocessing and normalization, invariantfeature extraction, and the advantages and disadvantages of these methods are analyzed.
2. An illumination invariant algorithm based on Nonsubsampled contourlet transform(NSCT) and NeighShrink denoise is proposed. Illumination is one of the factors that affectthe recognition efficiency. On the analysis of wavelet transform, we extract illuminationvariant feature through NSCT, which is a fully shift-invariant, multi-scale, multi-directiontransform and not only can extract more effective illumination invariant facial features butalso can get a clearer positive light image of face. A NeighShrink-based denoising model isapplied, which considers the correlations of sub-band coefficients. Thus, more usefulinformation can be restored in the high frequency sub-band coefficients, unlike some of the other approached in which too many coefficients that might contain useful informationtend to be killed. Experimental results showed that our method could extract invariantfeature more effective and obviously improve the recognition accuracy.
3. Face recognition system is divided into three key points, image preprocessing,feature extraction, classification and identification. Each part helps to improve the systemperformance. A variable light face recognition algorithm involved three points based onlocal binary pattern (LBP) and linear regression model is proposed. It can obtain betterperformance, through processing the illumination at each stage. In the image preprocessingphase, an efficient preprocessing chain is adopted which contains Gamma correction,difference of Gaussian(DoG) filtering, contrast equalization and can eliminates most of theeffects of changing illumination. Then block LBP is applied to extract invariant featurewhich is robust to variant illumination. Finally, improved linear regression model is used toclassification which drops the first n principal components. The proposed approach canreduce the effect of illumination, and improve robustness and the recognition accuracy offace recognition system.
本文针对光照问题进行研究，以提高人脸识别系统对光照的鲁棒性和识别率为主要目标，主要对人脸图像预处理、提取光照不变特征、分类识别等关键阶段展开研究，探讨人脸识别问题的研究方法。本文的主要工作和研究成果概括如下：
1.首先介绍了人脸识别课题的研究背景和意义及国内外发展情况，针对光照问题的三种研究方法：光照建模、光照补偿、光照不变特征提取法进行分类总结，分析其优缺点。
2.提出一种基于非下采样Contourlet变换（Nonsubsampled Contourlet Transform，NSCT）和邻域去噪的光照不变人脸识别算法。光照问题是影响识别效率的重要因素之一。本文在分析了小波变换提取人脸特征的基础上，采用多尺度的NSCT，它不仅具有小波变换的多分辨率和时频局域特性，还具有很强的方向性和冗余性，可以更完备的提取光照不变特征，同时在图像表示上能更好的描述人脸细节信息。采用邻域去噪方法去除光照不变特征中的投射阴影等噪声，因为投射阴影出现在局部的可能性最大，邻域去噪符合这点，其只在小范围内进行去噪，相比于一般的去噪方法，能保留更多的边界信息。经实验证明，该方法有效的提取光照不变特征，显著提高了人脸识别率。
3.人脸识别系统分为预处理、特征提取、分类识别三个关键环节，每一阶段对人脸识别系统性能都有所提升。本文提出的基于局部二值模式（Local Binary Pattern，LBP）和线性回归的可变光照人脸识别算法同时涉及这三个阶段，在每一阶段中对光照进行处理，得到更完备的可变光照算法。在预处理阶段采用Gamma校正、DoG滤波、对比度均衡化方法，降低光照敏感度，采用具有光照鲁棒性的分块LBP提取光照不变特征，最后使用改进的线性回归模型进行分类识别，即消除受光照影响最大的主成分系数。所提方法能有效的消除光照对人脸识别的影响，提高人脸识别系统的鲁棒性和识别率。
Face recognition is an important research topic in image processing and computervision, which has promising applications in public security, smart surveillance, multimediaand so on. Through face recognition has achieved great progress in the decades, when itcomes to uncontrolled conditions, such as different illumination conditions, pose variation,mixture of emotions and object shelter, and the accuracy of face recognition willdramatically decline. Therefore, how to build a real-life face recognition system is a sharpchallenging topic.
This paper focuses on the problem of face recognition in variant illuminationenvironment. The main purpose is to improve the recognition accuracy and robustness offace recognition system under various lighting condition. To achieve this purpose, thefocus of our work is on the image preprocessing, illumination invariant feature extraction,classification and recognition. The major work and contributions of this paper are follows:
1. The research background and significance of face recognition both at home andabroad are introduced firstly in this paper. Then the methods dealing with lighting problemare summarized, that is illumination model, preprocessing and normalization, invariantfeature extraction, and the advantages and disadvantages of these methods are analyzed.
2. An illumination invariant algorithm based on Nonsubsampled contourlet transform(NSCT) and NeighShrink denoise is proposed. Illumination is one of the factors that affectthe recognition efficiency. On the analysis of wavelet transform, we extract illuminationvariant feature through NSCT, which is a fully shift-invariant, multi-scale, multi-directiontransform and not only can extract more effective illumination invariant facial features butalso can get a clearer positive light image of face. A NeighShrink-based denoising model isapplied, which considers the correlations of sub-band coefficients. Thus, more usefulinformation can be restored in the high frequency sub-band coefficients, unlike some of the other approached in which too many coefficients that might contain useful informationtend to be killed. Experimental results showed that our method could extract invariantfeature more effective and obviously improve the recognition accuracy.
3. Face recognition system is divided into three key points, image preprocessing,feature extraction, classification and identification. Each part helps to improve the systemperformance. A variable light face recognition algorithm involved three points based onlocal binary pattern (LBP) and linear regression model is proposed. It can obtain betterperformance, through processing the illumination at each stage. In the image preprocessingphase, an efficient preprocessing chain is adopted which contains Gamma correction,difference of Gaussian(DoG) filtering, contrast equalization and can eliminates most of theeffects of changing illumination. Then block LBP is applied to extract invariant featurewhich is robust to variant illumination. Finally, improved linear regression model is used toclassification which drops the first n principal components. The proposed approach canreduce the effect of illumination, and improve robustness and the recognition accuracy offace recognition system.
引文
[1] Abate A F, Nappi M, Riccio D, et al.2D and3D face recognition: A survey[J].Pattern Recognition Letters,2007,28(14):1885-1906.
[2] Zhang X, Gao Y. Face recognition across pose: A review[J]. Pattern Recognition,2009,42(11):2876-2896.
[3] Phillips P J, Grother P, Micheals R J, et al. FRVT2002[J]. Evaluation Report, March,2003.
[4] Turk M A, Pentland A P. Face recognition using eigenfaces[C]. Computer Visionand Pattern Recognition,1991. Proceedings CVPR'91, IEEE Computer SocietyConference on. IEEE,1991:586-591.
[5] Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs. fisherfaces:Recognition using class specific linear projection[J]. Pattern Analysis and MachineIntelligence, IEEE Transactions on,1997,19(7):711-720.
[6] Georghiades A S, Belhumeur P N, Kriegman D J. From few to many: Illuminationcone models for face recognition under variable lighting and pose[J]. PatternAnalysis and Machine Intelligence, IEEE Transactions on,2001,23(6):643-660.
[7] Basri R, Jacobs D W. Lambertian reflectance and linear subspaces[J]. PatternAnalysis and Machine Intelligence, IEEE Transactions on,2003,25(2):218-233.
[8] Ramamoorthi R, Hanrahan P. On the relationship between radiance and irradiance:determining the illumination from images of a convex Lambertian object[J]. JOSAA,2001,18(10):2448-2459.
[9]胡元奎,汪增福.可变光照条件下的人脸图像识别[J].中国图象图形学报,2005,10(7):844-849.
[10] Savvides M, Kumar B V K V. Illumination normalization using logarithmtransforms for face authentication[C]. Audio-and Video-Based Biometric PersonAuthentication. Springer Berlin Heidelberg,2003:549-556.
[11] Shan S, Gao W, Cao B, et al. Illumination normalization for robust face recognitionagainst varying lighting conditions[C]. Analysis and Modeling of Faces andGestures,2003. AMFG2003. IEEE International Workshop on. IEEE,2003:157-164.
[12] Shashua A, Riklin-Raviv T. The quotient image: Class-based re-rendering andrecognition with varying illuminations[J]. Pattern Analysis and Machine Intelligence,IEEE Transactions on,2001,23(2):129-139.
[13] Land E H, McCann J J. Lightness and retinex theory[J]. Journal of the Opticalsociety of America,1971,61(1):1-11.
[14] Chen T, Yin W, Zhou X S, et al. Total variation models for variable lighting facerecognition[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on,2006,28(9):1519-1524.
[15]傅德胜,谢永华,任卫华.一种基于小波变换的分形图像编码压缩算法的研究[J].计算机应用与软件,2004,21(1):839-842.
[16]谢永华,陈伏兵,张生亮,等.融合小波变换与KPCA的分块人脸特征抽取与识别算法[J].中国图象图形学报,2007,12(4):666-672.
[17] Goh Y Z, Teoh A B J, Goh K O M. Wavelet-based illumination invariantpreprocessing in face recognition[J]. Journal of electronic imaging,2009,18(2):421-425.
[18] Zhang T, Fang B, Yuan Y, et al. Multiscale facial structure representation for facerecognition under varying illumination[J]. Pattern Recognition,2009,42(2):251-258.
[19] Cao X, Shen W, Yu L G, et al. Illumination invariant extraction for face recognitionusing neighboring wavelet coefficients[J]. Pattern Recognition,2012,45(4):1299-1305.
[20] Kittisuwan P, Asdornwised W. Wavelet-based image denoising using NeighShrinkand BiShrink threshold functions[C]. Electrical Engineering/Electronics, Computer,Telecommunications and Information Technology,2008. ECTI-CON2008.5thInternational Conference on. IEEE,2008,1:497-500.
[21] Do M N, Vetterli M. The contourlet transform: an efficient directionalmultiresolution image representation[J]. Image Processing, IEEE Transactions on,2005,14(12):2091-2106.
[22] Da Cunha A L, Zhou J, Do M N. The nonsubsampled contourlet transform: theory,design, and applications[J]. Image Processing, IEEE Transactions on,2006,15(10):3089-3101.
[23]贾建,陈莉.基于正态逆高斯模型的非下采样Contourlet变换图像去噪[J].电子学报,2011,39(7):1563-1568.
[24] Cheng Y, Hou Y, Zhao C, et al. Robust face recognition based on illuminationinvariant in nonsubsampled contourlet transform domain[J]. Neurocomputing,2010,73(10):2217-2224.
[25] Grother P, Micheals R J, Phillips P J. Face recognition vendor test2002performance metrics[C]. Audio-and Video-Based Biometric Person Authentication.Springer Berlin Heidelberg,2003:937-945.
[26] Harry Wechsler, P. Jonathon Philips, Vicki Bruce. Face Recognition from theory toapplication[M]. Springer in Cooperation with NATO Scientific Affairs Divison,2008.
[27] Mallat S G. A theory for multiresolution signal decomposition: the waveletrepresentation[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on,1989,11(7):674-693.
[28] Cooley J W, Tukey J W. An algorithm for the machine calculation of complexFourier series[J]. Mathematics of computation,1965,19(90):297-301.
[29] Garcia C, Zikos G, Tziritas G. A wavelet-based framework for face recognition[C].Int. Workshop on Advances in Facial Image Anal. Recognition Technology,5thEuropean Conf. Computer Vision.1998.
[30] Nastar C, Ayache N. Frequency-based nonrigid motion analysis: Application to fourdimensional medical images[J]. Pattern Analysis and Machine Intelligence, IEEETransactions on,1996,18(11):1067-1079.
[31] Zhang B L, Zhang H, Ge S S. Face recognition by applying wavelet subbandrepresentation and kernel associative memory[J]. Neural Networks, IEEETransactions on,2004,15(1):166-177.
[32] Candes E J, Donoho D L. Curvelets: A surprisingly effective nonadaptiverepresentation for objects with edges[R]. Stanford Univ Ca Dept of Statistics,2000.
[33]杨煊,裴继红,杨万海.小波变换方法在高分辨率多光谱图像融合中存在的问题[J].红外与毫米波学报,2002,21(1):77-80.
[34]贾建,焦李成,项海林.基于双变量阈值的非下采样Contourlet变换图像去噪[J].电子与信息学报,2009,31(3):532-536.
[35] Huang L, Wang H, Zhu B. Adaptive thresholds algorithm of image denoising basedon nonsubsampled contourlet transform[C]. Computer Science and SoftwareEngineering,2008International Conference on. IEEE,2008,6:209-212.
[36]雷浩鹏,李峰.基于多小波-非采样Contourlet变换的自适应阈值图像去噪方法[J].计算机应用,2010,30(5):1351-1355.
[37]张林,朱兆达.基于非降采样Contourlet变换的非线性图像增强新算法[J].电子与信息学报,2009,31(8):1786-1790.
[38] Li T, Wang Y. Biological image fusion using a NSCT based variable-weightmethod[J]. Information Fusion,2011,12(2):85-92.
[39] Yang X H, Jiao L C. Fusion algorithm for remote sensing images based onnonsubsampled contourlet transform[J]. Acta Automatica Sinica,2008,34(3):274-281.
[40] Ojala T, Pietik inen M, Harwood D. A comparative study of texture measures withclassification based on featured distributions[J]. Pattern recognition,1996,29(1):51-59.
[41] Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray-scale and rotationinvariant texture classification with local binary patterns[J]. Pattern Analysis andMachine Intelligence, IEEE Transactions on,2002,24(7):971-987.
[42] Horn B K P. Robot vision[M]. The MIT Press,1986.
[43] Gross R, Brajovic V. An image preprocessing algorithm for illumination invariantface recognition[C]. Audio-and Video-Based Biometric Person Authentication.Springer Berlin Heidelberg,2003:10-18.
[44] Sim T, Baker S, Bsat M. The CMU pose, illumination, and expression (PIE)database[C]. Automatic Face and Gesture Recognition,2002. Proceedings. FifthIEEE International Conference on. IEEE,2002:46-51.
[45] Adini Y, Moses Y, Ullman S. Face recognition: The problem of compensating forchanges in illumination direction[J]. Pattern Analysis and Machine Intelligence,IEEE Transactions on,1997,19(7):721-732.
[46] R. Sokal, F. Rohlf. Biometry [M]. Freeman: San Francisco,1969.
[47] Naseem I, Togneri R, Bennamoun M. Linear regression for face recognition[J].Pattern Analysis and Machine Intelligence, IEEE Transactions on,2010,32(11):2106-2112.
[48] Naseem I, Togneri R, Bennamoun M. Robust regression for face recognition[J].Pattern Recognition,2012,45(1):104-118.
[49] Huang S M, Yang J F. Improved principal component regression for facerecognition under illumination variations[J]. Signal Processing Letters, IEEE,2012,19(4):179-182.
[50] Huang S M, Yang J F. Kernel linear regression for low resolution face recognitionunder variable illumination[C]. Acoustics, Speech and Signal Processing (ICASSP),2012IEEE International Conference on. IEEE,2012:1945-1948.
[51] Tahir M A, Chan C H, Kittler J, et al. Face recognition using multi-scale local phasequantisation and Linear Regression Classifier[C]. Image Processing (ICIP),201118th IEEE International Conference on. IEEE,2011:765-768.
[52] Chai X, Shan S, Chen X, et al. Locally linear regression for pose-invariant facerecognition[J]. Image Processing, IEEE Transactions on,2007,16(7):1716-1725.
[53] Hastie T, Tibshirani R, Friedman J, et al. The elements of statistical learning: datamining, inference and prediction[J]. The Mathematical Intelligencer,2005,27(2):83-85.
[54] Seber G A F, Lee A J. Linear regression analysis[M]. Wiley,2012.
[55] Ryan T P. Modern regression methods[M]. Wiley-Interscience,2008.
[56] A. Martinez and R. Benavente. The AR Face Database. CVC Technical Report24,1998.
[57] Phillips P J, Wechsler H, Huang J, et al. The FERET database and evaluationprocedure for face-recognition algorithms[J]. Image and vision computing,1998,16(5):295-306.
[2] Zhang X, Gao Y. Face recognition across pose: A review[J]. Pattern Recognition,2009,42(11):2876-2896.
[3] Phillips P J, Grother P, Micheals R J, et al. FRVT2002[J]. Evaluation Report, March,2003.
[4] Turk M A, Pentland A P. Face recognition using eigenfaces[C]. Computer Visionand Pattern Recognition,1991. Proceedings CVPR'91, IEEE Computer SocietyConference on. IEEE,1991:586-591.
[5] Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs. fisherfaces:Recognition using class specific linear projection[J]. Pattern Analysis and MachineIntelligence, IEEE Transactions on,1997,19(7):711-720.
[6] Georghiades A S, Belhumeur P N, Kriegman D J. From few to many: Illuminationcone models for face recognition under variable lighting and pose[J]. PatternAnalysis and Machine Intelligence, IEEE Transactions on,2001,23(6):643-660.
[7] Basri R, Jacobs D W. Lambertian reflectance and linear subspaces[J]. PatternAnalysis and Machine Intelligence, IEEE Transactions on,2003,25(2):218-233.
[8] Ramamoorthi R, Hanrahan P. On the relationship between radiance and irradiance:determining the illumination from images of a convex Lambertian object[J]. JOSAA,2001,18(10):2448-2459.
[9]胡元奎,汪增福.可变光照条件下的人脸图像识别[J].中国图象图形学报,2005,10(7):844-849.
[10] Savvides M, Kumar B V K V. Illumination normalization using logarithmtransforms for face authentication[C]. Audio-and Video-Based Biometric PersonAuthentication. Springer Berlin Heidelberg,2003:549-556.
[11] Shan S, Gao W, Cao B, et al. Illumination normalization for robust face recognitionagainst varying lighting conditions[C]. Analysis and Modeling of Faces andGestures,2003. AMFG2003. IEEE International Workshop on. IEEE,2003:157-164.
[12] Shashua A, Riklin-Raviv T. The quotient image: Class-based re-rendering andrecognition with varying illuminations[J]. Pattern Analysis and Machine Intelligence,IEEE Transactions on,2001,23(2):129-139.
[13] Land E H, McCann J J. Lightness and retinex theory[J]. Journal of the Opticalsociety of America,1971,61(1):1-11.
[14] Chen T, Yin W, Zhou X S, et al. Total variation models for variable lighting facerecognition[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on,2006,28(9):1519-1524.
[15]傅德胜,谢永华,任卫华.一种基于小波变换的分形图像编码压缩算法的研究[J].计算机应用与软件,2004,21(1):839-842.
[16]谢永华,陈伏兵,张生亮,等.融合小波变换与KPCA的分块人脸特征抽取与识别算法[J].中国图象图形学报,2007,12(4):666-672.
[17] Goh Y Z, Teoh A B J, Goh K O M. Wavelet-based illumination invariantpreprocessing in face recognition[J]. Journal of electronic imaging,2009,18(2):421-425.
[18] Zhang T, Fang B, Yuan Y, et al. Multiscale facial structure representation for facerecognition under varying illumination[J]. Pattern Recognition,2009,42(2):251-258.
[19] Cao X, Shen W, Yu L G, et al. Illumination invariant extraction for face recognitionusing neighboring wavelet coefficients[J]. Pattern Recognition,2012,45(4):1299-1305.
[20] Kittisuwan P, Asdornwised W. Wavelet-based image denoising using NeighShrinkand BiShrink threshold functions[C]. Electrical Engineering/Electronics, Computer,Telecommunications and Information Technology,2008. ECTI-CON2008.5thInternational Conference on. IEEE,2008,1:497-500.
[21] Do M N, Vetterli M. The contourlet transform: an efficient directionalmultiresolution image representation[J]. Image Processing, IEEE Transactions on,2005,14(12):2091-2106.
[22] Da Cunha A L, Zhou J, Do M N. The nonsubsampled contourlet transform: theory,design, and applications[J]. Image Processing, IEEE Transactions on,2006,15(10):3089-3101.
[23]贾建,陈莉.基于正态逆高斯模型的非下采样Contourlet变换图像去噪[J].电子学报,2011,39(7):1563-1568.
[24] Cheng Y, Hou Y, Zhao C, et al. Robust face recognition based on illuminationinvariant in nonsubsampled contourlet transform domain[J]. Neurocomputing,2010,73(10):2217-2224.
[25] Grother P, Micheals R J, Phillips P J. Face recognition vendor test2002performance metrics[C]. Audio-and Video-Based Biometric Person Authentication.Springer Berlin Heidelberg,2003:937-945.
[26] Harry Wechsler, P. Jonathon Philips, Vicki Bruce. Face Recognition from theory toapplication[M]. Springer in Cooperation with NATO Scientific Affairs Divison,2008.
[27] Mallat S G. A theory for multiresolution signal decomposition: the waveletrepresentation[J]. Pattern Analysis and Machine Intelligence, IEEE Transactions on,1989,11(7):674-693.
[28] Cooley J W, Tukey J W. An algorithm for the machine calculation of complexFourier series[J]. Mathematics of computation,1965,19(90):297-301.
[29] Garcia C, Zikos G, Tziritas G. A wavelet-based framework for face recognition[C].Int. Workshop on Advances in Facial Image Anal. Recognition Technology,5thEuropean Conf. Computer Vision.1998.
[30] Nastar C, Ayache N. Frequency-based nonrigid motion analysis: Application to fourdimensional medical images[J]. Pattern Analysis and Machine Intelligence, IEEETransactions on,1996,18(11):1067-1079.
[31] Zhang B L, Zhang H, Ge S S. Face recognition by applying wavelet subbandrepresentation and kernel associative memory[J]. Neural Networks, IEEETransactions on,2004,15(1):166-177.
[32] Candes E J, Donoho D L. Curvelets: A surprisingly effective nonadaptiverepresentation for objects with edges[R]. Stanford Univ Ca Dept of Statistics,2000.
[33]杨煊,裴继红,杨万海.小波变换方法在高分辨率多光谱图像融合中存在的问题[J].红外与毫米波学报,2002,21(1):77-80.
[34]贾建,焦李成,项海林.基于双变量阈值的非下采样Contourlet变换图像去噪[J].电子与信息学报,2009,31(3):532-536.
[35] Huang L, Wang H, Zhu B. Adaptive thresholds algorithm of image denoising basedon nonsubsampled contourlet transform[C]. Computer Science and SoftwareEngineering,2008International Conference on. IEEE,2008,6:209-212.
[36]雷浩鹏,李峰.基于多小波-非采样Contourlet变换的自适应阈值图像去噪方法[J].计算机应用,2010,30(5):1351-1355.
[37]张林,朱兆达.基于非降采样Contourlet变换的非线性图像增强新算法[J].电子与信息学报,2009,31(8):1786-1790.
[38] Li T, Wang Y. Biological image fusion using a NSCT based variable-weightmethod[J]. Information Fusion,2011,12(2):85-92.
[39] Yang X H, Jiao L C. Fusion algorithm for remote sensing images based onnonsubsampled contourlet transform[J]. Acta Automatica Sinica,2008,34(3):274-281.
[40] Ojala T, Pietik inen M, Harwood D. A comparative study of texture measures withclassification based on featured distributions[J]. Pattern recognition,1996,29(1):51-59.
[41] Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray-scale and rotationinvariant texture classification with local binary patterns[J]. Pattern Analysis andMachine Intelligence, IEEE Transactions on,2002,24(7):971-987.
[42] Horn B K P. Robot vision[M]. The MIT Press,1986.
[43] Gross R, Brajovic V. An image preprocessing algorithm for illumination invariantface recognition[C]. Audio-and Video-Based Biometric Person Authentication.Springer Berlin Heidelberg,2003:10-18.
[44] Sim T, Baker S, Bsat M. The CMU pose, illumination, and expression (PIE)database[C]. Automatic Face and Gesture Recognition,2002. Proceedings. FifthIEEE International Conference on. IEEE,2002:46-51.
[45] Adini Y, Moses Y, Ullman S. Face recognition: The problem of compensating forchanges in illumination direction[J]. Pattern Analysis and Machine Intelligence,IEEE Transactions on,1997,19(7):721-732.
[46] R. Sokal, F. Rohlf. Biometry [M]. Freeman: San Francisco,1969.
[47] Naseem I, Togneri R, Bennamoun M. Linear regression for face recognition[J].Pattern Analysis and Machine Intelligence, IEEE Transactions on,2010,32(11):2106-2112.
[48] Naseem I, Togneri R, Bennamoun M. Robust regression for face recognition[J].Pattern Recognition,2012,45(1):104-118.
[49] Huang S M, Yang J F. Improved principal component regression for facerecognition under illumination variations[J]. Signal Processing Letters, IEEE,2012,19(4):179-182.
[50] Huang S M, Yang J F. Kernel linear regression for low resolution face recognitionunder variable illumination[C]. Acoustics, Speech and Signal Processing (ICASSP),2012IEEE International Conference on. IEEE,2012:1945-1948.
[51] Tahir M A, Chan C H, Kittler J, et al. Face recognition using multi-scale local phasequantisation and Linear Regression Classifier[C]. Image Processing (ICIP),201118th IEEE International Conference on. IEEE,2011:765-768.
[52] Chai X, Shan S, Chen X, et al. Locally linear regression for pose-invariant facerecognition[J]. Image Processing, IEEE Transactions on,2007,16(7):1716-1725.
[53] Hastie T, Tibshirani R, Friedman J, et al. The elements of statistical learning: datamining, inference and prediction[J]. The Mathematical Intelligencer,2005,27(2):83-85.
[54] Seber G A F, Lee A J. Linear regression analysis[M]. Wiley,2012.
[55] Ryan T P. Modern regression methods[M]. Wiley-Interscience,2008.
[56] A. Martinez and R. Benavente. The AR Face Database. CVC Technical Report24,1998.
[57] Phillips P J, Wechsler H, Huang J, et al. The FERET database and evaluationprocedure for face-recognition algorithms[J]. Image and vision computing,1998,16(5):295-306.