基于局部纹理特征融合的面部表情识别方法研究
|
摘要
本文从表情识别的局部纹理信息分析入手，主要研究人脸关键点（如人眼）的定位、通过定位关键点（如人眼）来完成标准人脸图像的获取，进而将人脸面部表情放在局部特征的背景下进行识别，旨在达到两个目的：一是提高表情识别的准确率；二是提高算法的鲁棒性。此外，本论文还研究了一种新的基于决策树算法的人脸面部表情分类方法，提高了整个人脸面部表情识别的分类速度,为人脸面部表情识别分类研究提供了一种全新分析数据、预测数据的手段。同时，针对传统支持向量机分类参数采用网格寻优的方式，提出了一种利用粒子群算法对其进行自动优化的方法，提高了整个人脸面部表情识别系统分类的效率。本文主要针对静态人脸面部表情图像的以下方面进行了研究与探讨：如何定位人脸的关键部位、如何对人脸进行检测、人脸面部表情特征的提取与表征、如何对人脸面部表情进行分类。本研究的具体内容如下：
1.提出了一种新颖高效的由粗到精的人眼检测策略，应用Gabor变换和PCA重构验证来粗略定位眼睛区域，再使用两级邻域运算对瞳孔进行精确定位。该方法具有非迭代和计算简单的优点，通过使用显著极值检测及PCA重构验证，省去训练样本的步骤，从而在保证定位精度的同时大大降低了计算量。实验结果表明，该方法的眼睛定位准确率可达到98.6%，实现了对眼睛的精确定位。
2.采用Gabor滤波器提取人脸面部表情图像的多尺度和多方向特征,按照融合规则将相同尺度不同方向的Gabor模特征融合到一起。并利用LBP算子对Gabor模特征进一步编码，为了能够有效地表征图像全局特征,将采用子块分割法，即将融合图像分割为多个等面积且互相之间不重叠的小单元，对每一个单元内部的融合特征画出直方图，将所有的直方图结合起来分析表情图像。经实验验证，通过该方法提取的特征所得到的人脸面部表情识别率可达到93.42%，无论在计算量上还是识别性能上都比传统的特征提取更具有优势。
3.针对传统支持向量机分类参数采用网格寻优的方式，提出了一种利用粒子群算法对其进行自动优化的方法，通过优化明确了最优的参数选择，提高了支持向量机的性能，经实验验证，该种优化的方法可以将识别率提高至96.05%，证明了此方法的有效性，提高了整个人脸面部表情识别系统分类的效率。此外，本文还将基于决策树算法运用到人脸面部表情分类中。不仅可以将其作为对比本文分类实验结果的一个依据而且还为人脸面部表情识别分类研究提供了一种全新分析数据、预测数据的手段。
4.对全文所做的工作进行了总结，对文章中存在的不足以及今后需要研究问题的进行了说明。
论文研究得到吉林省科技发展计划重点项目“基于混合特征的人脸表情识别方法与系统研究”（20071152），吉林大学“2010年研究生创新研究计划项目”（20101027）及吉林省青年科研基金项目“局部遮挡情况下的鲁棒表情识别方法研究”（20140520065JH）的资助。
This paper studied the local texture feature of the expression recognition and the locationof key feature (such as eyes) was investigated on facial expression recognition in order torecognize the facial expressions based on the local feature. The purpose of this research is toimprove the facial recognition and the robustness of the algorithm. A novel facial expressionclassification based on the decision tree algorithm was provided in this paper, whichimproved the classification of facial expression recognition and provided an effectivemethod of data analysis and forecasts. The particle swarm optimization (PSO) algorithm wasused to solve the problem of traditional support vector machine (SVM) in order to improvethe accuracy of facial expression recognition system. For the static facial expression images,the location of the key facial feature, the face detection, the extraction and expression of thefeature of facial expression, and the classification were studied in this research. The mainresearch and innovative work are shown as follows:
First, Gabor transform and principal component analysis (PCA) refactoring validationwere used to locate the eye area roughly and the two stage neighborhood arithmetic was usedto locate the pupil precisely. This method has the advantages of the non-iteratively and thesimple computation. Salient image extrema detection and PCA refactoring validation wereused to reduce the step of training sample computation under the requirement of locationaccuracy. The accuracy of eye location was more than98.6%as showed in the experimentresult.
Second, Gabor filterbank was provided to extract the multi-scale and multi-directioncharacteristic of facial expression images. The features on the same scale in differentdirections of Gabor model syndrome were fused based on the fusion rules. The local binarypattern (LBP) operator was used to encode the Gabor model syndrome in. The sub-locksegmentation was used to improve the recognition of global image features. The image wassegmented into non-superposed elements with the equal area. Histograms of every element’sfusion features were combined to analyse the express image. The accuracy of facial expression recognition was93.42%as showed in the experiment. It has the advantage incomputation and recognition property compared with traditional method.
Third, the PSO algorithm was provided in the traditional support vector machine withgrid search technique of classifying parameter to improve the efficiency of facial expressionrecognition system. The accuracy of recognition was improved to96.05%.Thedecision tree algorithm was used in classification of facial expression in this research whichprovided comparation of classification experiment and an effective method of data analysisand forecasts.
Last, the main work of this research is summarized and the further work is discussed.
This work was supported by Jilin Provincial Science and Technology development plankey Program (No.20071152), Graduate innovation project of Jilin University (No.20101027)and Jilin Youth Scientific Fund Project (No.20140520065JH).
1.提出了一种新颖高效的由粗到精的人眼检测策略，应用Gabor变换和PCA重构验证来粗略定位眼睛区域，再使用两级邻域运算对瞳孔进行精确定位。该方法具有非迭代和计算简单的优点，通过使用显著极值检测及PCA重构验证，省去训练样本的步骤，从而在保证定位精度的同时大大降低了计算量。实验结果表明，该方法的眼睛定位准确率可达到98.6%，实现了对眼睛的精确定位。
2.采用Gabor滤波器提取人脸面部表情图像的多尺度和多方向特征,按照融合规则将相同尺度不同方向的Gabor模特征融合到一起。并利用LBP算子对Gabor模特征进一步编码，为了能够有效地表征图像全局特征,将采用子块分割法，即将融合图像分割为多个等面积且互相之间不重叠的小单元，对每一个单元内部的融合特征画出直方图，将所有的直方图结合起来分析表情图像。经实验验证，通过该方法提取的特征所得到的人脸面部表情识别率可达到93.42%，无论在计算量上还是识别性能上都比传统的特征提取更具有优势。
3.针对传统支持向量机分类参数采用网格寻优的方式，提出了一种利用粒子群算法对其进行自动优化的方法，通过优化明确了最优的参数选择，提高了支持向量机的性能，经实验验证，该种优化的方法可以将识别率提高至96.05%，证明了此方法的有效性，提高了整个人脸面部表情识别系统分类的效率。此外，本文还将基于决策树算法运用到人脸面部表情分类中。不仅可以将其作为对比本文分类实验结果的一个依据而且还为人脸面部表情识别分类研究提供了一种全新分析数据、预测数据的手段。
4.对全文所做的工作进行了总结，对文章中存在的不足以及今后需要研究问题的进行了说明。
论文研究得到吉林省科技发展计划重点项目“基于混合特征的人脸表情识别方法与系统研究”（20071152），吉林大学“2010年研究生创新研究计划项目”（20101027）及吉林省青年科研基金项目“局部遮挡情况下的鲁棒表情识别方法研究”（20140520065JH）的资助。
This paper studied the local texture feature of the expression recognition and the locationof key feature (such as eyes) was investigated on facial expression recognition in order torecognize the facial expressions based on the local feature. The purpose of this research is toimprove the facial recognition and the robustness of the algorithm. A novel facial expressionclassification based on the decision tree algorithm was provided in this paper, whichimproved the classification of facial expression recognition and provided an effectivemethod of data analysis and forecasts. The particle swarm optimization (PSO) algorithm wasused to solve the problem of traditional support vector machine (SVM) in order to improvethe accuracy of facial expression recognition system. For the static facial expression images,the location of the key facial feature, the face detection, the extraction and expression of thefeature of facial expression, and the classification were studied in this research. The mainresearch and innovative work are shown as follows:
First, Gabor transform and principal component analysis (PCA) refactoring validationwere used to locate the eye area roughly and the two stage neighborhood arithmetic was usedto locate the pupil precisely. This method has the advantages of the non-iteratively and thesimple computation. Salient image extrema detection and PCA refactoring validation wereused to reduce the step of training sample computation under the requirement of locationaccuracy. The accuracy of eye location was more than98.6%as showed in the experimentresult.
Second, Gabor filterbank was provided to extract the multi-scale and multi-directioncharacteristic of facial expression images. The features on the same scale in differentdirections of Gabor model syndrome were fused based on the fusion rules. The local binarypattern (LBP) operator was used to encode the Gabor model syndrome in. The sub-locksegmentation was used to improve the recognition of global image features. The image wassegmented into non-superposed elements with the equal area. Histograms of every element’sfusion features were combined to analyse the express image. The accuracy of facial expression recognition was93.42%as showed in the experiment. It has the advantage incomputation and recognition property compared with traditional method.
Third, the PSO algorithm was provided in the traditional support vector machine withgrid search technique of classifying parameter to improve the efficiency of facial expressionrecognition system. The accuracy of recognition was improved to96.05%.Thedecision tree algorithm was used in classification of facial expression in this research whichprovided comparation of classification experiment and an effective method of data analysisand forecasts.
Last, the main work of this research is summarized and the further work is discussed.
This work was supported by Jilin Provincial Science and Technology development plankey Program (No.20071152), Graduate innovation project of Jilin University (No.20101027)and Jilin Youth Scientific Fund Project (No.20140520065JH).
引文
[1]金辉，高文.基于HMM的面部表情图像序列的分析与识别[J].自动化学报，2002，28(4):646-650.
[2] Lyons M, Akamatsu S, Kamachi M, et al. Coding facial expressions with Gaborwavelets[C]. Proceeding of the Thrid IEEE International Conference on Automatic Faceand Gesture Recognition. Nara, Japan,1998:200-205.
[3] Wong S K and Ziarko W. On optimal decision rules in decision tables[J]. Bulletin ofPolish Academy of Science,1985,33:693696.
[4] Mikolajczyk K, Tuytelaars T, Schmid C, et al. A comparison of affine regiondetectors[J]. International Journal of Computer Vision,2005,65(1-2):43-72.
[5] Belhumeur P N, Hespanha J, Kriegman D J. Eigenfaces vs. fisherfaces: recognitionusing class specific linear projection[J]. IEEE Pattern Analysis and MachineIntelligence,1997,19(7):711-720.
[6] Huang R, Liu Q and Lu H. Solving the small sample size problem of LDA[C]. InProceeding of16th International Conference on Pattern Recognition (ICPR2002).Quebec, Canada: IEEE Computer Society,2002:29-32.
[7] Dai D Q and Pong C. Yuenv. Face recognition by regularized discriminant analysis[J].Systems, Man, and Cybernetics part B-cybernetics,2007,37(4):1080-1085.
[8] Mendes R, Kenney J. The fully informed particle swarm: Simpler, maybe better[J].IEEE Transaction on Evolutionary Computation,2004,8(3):204-210.
[9] Ye J and Li Q. A two-stage linear discriminant analysis via QR-decomposition[J].Pattern Analysis and Machine Intelligence,2005,27(6):929-941.
[10] Chen L F, Liao H M and Ko M T. A new LDA-based face recognition system whichcan solve the small sample size problem[J]. Pattern Recognition,2000,33(10):1713-1726.
[11] Liu J and Chen S. Discriminant common vecotors versus neighbourhood componentsanalysis and laplacianfaces: A comparative study in small sample size problem[J].Image and Vision Computing,2006,24(3):249-262.
[12] Wang X and Tang X. Random sampling for subspace face recognition[J]. InternationalJournal of Computer Vision,2006,70(1):94-104.
[13] Chawla N V and Bowyer K W. Random subspaces and subsampling for2-D facerecognition[C]. Computer Vision and Pattern Recognition (CVPR2005). San Diego,USA: IEEE Computer Society,2005:582-589.
[14] Wang X, Huang C and Fang X.2DPCA vs.2DLDA: face recognition usingtwo-dimensional method[C]. Artificial Intelligence and Computational Intelligence.Shanghai, China: Springer,2009:357-360.
[15] Y Jian, Zhang D, Yong X and Yang J. Two-dimensional discriminant transform forface recognition[J]. Pattern Recognition, July2005,38(7):1125-1129.
[16] Noushath S, Hemantha Kumar G and Shivakumara P.(2D)2LDA: An efficientapproach for face recognition[J]. Pattern Recognition,2006,39(7):1369-1400.
[17] Luo D, Ding C and Huang H. Symmetric two dimensional linear discriminantanalysis(2DLDA)[C]. Computer Vision and Pattern Recognition (CVPR2009). Miami, FL,USA: IEEE Computer Society,2009:2812-2819.
[18] Wang X, Huang C and Liu J. Gabor-2DLDA: face recognition using Gabor featuresand2D linear discriminant analysis[C]. Computation Technology and Automation.Changsha, China: IEEE Computer Society,2009:608-610.
[19] Wang J, Lu H, Plataniotis K N and Lu J. Gaussian kernel optimization for patternclassification[J]. Pattern Recognition, July2009,42(7):1237-1247.
[20] Wang J, Plataniotis K N, Lu J and Venetsanopoulos A N. Kernel QuadraticDiscriminant analysis for small sample size problem[J]. Pattern Recognition,2008,39(5):1528-1538.
[21] Lu J, Plataniotis K N and Venetsanopoulos A N. Kernel discriminant learning withapplication to face recognition[C]. Support Vector Machines: Theory and Applications.Berlin Heidelberg: Springer-Verlag,2005:275-296.
[22] Lu J, Plataniotis K N, Venetsanopoulos A N and Wang J. An efficient kerneldiscriminant analysis method[J]. Pattern Recognition,2005,38(10):1788-1790.
[23] Lu J, Plataniotis K N and Venetsanopoulos A N. Face recognition using kernel directdiscriminant analysis algorithms[J]. IEEE Transactions on Neural Networks, January2003,14(1):117-126.
[24] Wang J, Plataniotis K N, Lu J and Venetsanopoulos A N. On solving the facerecognition problem with one training sample per subject[J]. Pattern Recognition,2006,39(9):1746-1762.
[25] Liu J, Chen S and Zhou Z. Single image subspace for face Recognition[C]. Analysisand modeling of faces and gestures. Rio de Janeiro, Brazil: Springer-Verlag Berlin,Heidelberg,2007:205-219.
[26] Chen S, Liu J and Zhou Z. Making FLDA application to face recognition with onesample per person[J]. Pattern Recognition,2004,37(7):1553-1555.
[27] Zheng W, Zhao L and Zou C. An efficient algorithm to solve the small sample sizeproblem for LDA[J]. Pattern Recognition, May2004,37(5):1077-1079.
[28] Zhou J, Lu C, Zhang C and Li Y. A survey of automatic human face recognition[J].Acta Electronica Sinica,2000,28(4):102-106.
[29] Tao D, Tang X, Li X and Rui Y. Kernel direct biased discriminant analysis: a newcontent-based image retrieval relevance feedback algorithm[J]. IEEE Transactions onMultimedia,2006,8:716-727.
[30] Philbin J, Chum O, Isard M, Sivic J and Zisserman A. Object retrieval with largevocabularies and fast spatial matching[C]. Proceedings of the IEEE Conference onComputer Vision and Pattern Recognition,2007:1-8.
[31] Schaffalitzky F and Zisserman A. Multi-view matching for unordered image sets[M].Proceedings of the European Conference on Computer Vision,2002:414-431.
[32] Tuytelaars T and Van Gool L. Matching widely separated views based on affineinvariant regions[J]. International Journal of Computer Vision,2004,59(1):61-85.
[33] Ferrari V, Tuytelaars T and Van Gool L. Simultaneous object recognition andsegmentation by image exploration[M]. Proceedings of the European Conference onComputer Vision,2004:40-54.
[34] Lazebnik S, Schmid C and Ponce J. Sparse texture representation usingaffine-invariant neighborhoods[C]. Proceedings of the IEEE Conference on ComputerVision and Pattern Recognition,2003:319-324.
[35] Fergus R, Perona P and Zisserman A. Object class recognition by unsupervisedscale-invariant learning[C]. Proceedings of the IEEE Conference on Computer Visionand Pattern Recognition,2003,2:264-271
[36] Brown M and Lowe D. Recognising panoramas[C]. Proceedings of InternationalConference on Computer Vision,2003,3:1218-1227.
[37] Ferrari V, Tuytelaars T and Van Gool L. Wide-baseline multiple-viewcorrespondences[C]. Proceedings of the IEEE Conference on Computer Vision andPattern Recognition,2003,1:718-725.
[38]王永明，王贵锦.图像局部不变性特征与描述[B]，北京：国防工业出版，2010.
[39] Moravec H P. Towards automatic visual obstacle avoidance[C]. Proceedings ofInternational Conference on Artificial Intelligence,1977:584.
[40] Harris C and Stephens M. A combined corner and edge detector[C]. Alvey VisionConference,1988:147-151.
[41] Smith S M and Brady J M. SUSAN-a new approach to low level image[J].Processing, International Journal of Computer Vision,1997,23(1):45-78.
[42] Matas J, Chum O, Urban M and Pajdla T. Robust wide-baseline stereo frommaximally stable extremal regions[J]. Proceedings of British Machine VisionConference,2004,22(10):761-767.
[43] Lowe D G.. Distinctive image features from scale-invariant keypoints[J]. InternationalJournal of Computer Vision,2004,60(2):91-110.
[44] Mokhtarian F and Suomela R. Robust image corner detection through curvaturescale-space[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(12):1376–1381.
[45] Mikolajczyk K and Schmid C. Scale and affine invariant interest point detectors[J].International Journal on Computer Vision,2004,60(1):63-86.
[46] Tuytelaars T and Van Gool L. Matching widely separated views based on affineinvariant regions[J]. International Journal of Computer Vision,2004,59(1):61-85.
[47] Matas J, Chum O, Urban M and Pajdla T. Robust wide-baseline stereo frommaximally stable extremal regions[J]. Proceedings of British Machine VisionConference,2004,22(10):761-767.
[48] Kadir T, Zisserman A and Brady M. An affine invariant salient region detector[M].Proceedings of the European Conference on Computer Vision,2004,3021:228-241.
[49] Mikolajczyk K and Schmid C. A performance evaluation of local descriptors[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2005,27(10):1615-1630.
[50] Ma L and Khorasani K. Facial expression recognition using constructive feedforwardNeural Networks[J]. IEEE Transactions on Systems Man and Cybernetics,2004,34(3):1588-1595.
[51] Gueorguieva N, Georgiev G and Valova I. Facial expression recognition usingfeedforward neural networks[C]. Proceedings of the International Conference onArtificial Intelligence, Las Vegas, N V, USA,2003:285-291.
[52] Soyel H and Demirel H. Facial expression recognition using3D facial featuredistances[C]. Proceedings of the4th International Conference on Image Analysis andRecognition, Montreal, Canada,2007:831-838.
[53]周书仁，梁昔明，杨秋芬.类间学习神经网络的人脸表情识别[J].计算机应用研究，2008，25(7)：2219-2222.
[54] Sebe N, Cohen I, Garg A, et al. Emotion recognition using a Cauchy naive BayesClassifier[C]. Proceedings of International Conference on Pattern Recognition,Québec City, Canada,2002,(1):17-20.
[55] Cohen I, Sebe N, Garg A, et al. Facial expression recognition from video sequences:Temporal and staticmodeling[J]. Computer Vision and Image Understanding,2003,91(1-2):160-187.
[56] Cohen I, Sebe N, Cozman F G, et al. Learning bayesian network classifiers for facialexpression recognition with both labeled and unlabeled data[C]. Proceedings ofInternational Conference on Computer Vision and Pattern Recognition, Madison,Wisconsin, USA,2003,(1):595-604.
[57] Zhang Y and Ji Q. Active and dynamic information fusion for facial expressionunderstanding from image sequences[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,2005,27(5):699-714.
[58]朴林植，赵杰煜.基于统计形状分析的人脸基本表情分析[J].宁波大学学报(理工版),2009，22(2)：196-210.
[59] Littlewort G, Bartlett M S, Fasel I, et al. Analysis of machine learning methods forreal-time recognition of facial expressions from video[C]. Machine PerceptionLaboratory Institute for Neural Computation, University of California, San Diego, CA,USA,2004:27.
[60] Xu Q Z, Zhang P Z, Yang L X, et al. A facial expression recognition approach basedon novel support vector machine tree[C]. Proceedings of the4th InternationalSymposium on Neural Networks, Nanjing, China,2007:374-381.
[61] Bartlett M S, Littlewort G, Frank M, et al. Recognizing facial expression: machinelearning and application to spontaneous behavior[C]. Proceedings of IEEE ComputerSociety Conference on Computer Vision and Pattern Recognition, San Diego, CA,USA,2005,2:568-573.
[62]孙雯玉.人脸表情识别算法研究[D].北京交通大学硕士学位论文,2007.
[63] Muller S, Wallhoff F, Hulsken F, et al. Facial expression recognition using pseudo3-Dhidden Markov models[C]. Proceedings of International Conference on PatternRecognition, Québec City, Canada,2002,2:32~35.
[64] Yeasin M, Bullot B and Sharma R. From facial expression to level of interest: aspatio-temporal approach[C]. Proceedings of International Conference on ComputerVision and Pattern Recognition, Washington, DC, USA,2004,2:922-927.
[65] Shan S, Chang Y and Gao W. Curse of mis-alignment in face recognition: problem anda novel mis-alignment learning solution[C]. Proeedings of International Conference onAutomatic Face and Gesture Recognition,2004,314–320.
[66] Wang P, Green M, Ji Q and Waymanm J. Automatic eye detection and its validation[C].Proeedings of International Conference on Computer Vision and Pattern Recognition,2005,164–164.
[67] Rodriguez Y, Cardinaux F, Bengio S and Marithoz J. Measuring the performance offace localization systems[J]. Image Vision Comput,2006,24(8):882-893.
[68] Haro A, Flickner M and Essa I. Detecting and tracking eyes by using theirphysiological properties, dynamics, and appearance[C]. Proeedings of InternationalConference on Computer Vision and Pattern Recognition,2000,163-168.
[69] D’Orazio T, Leo M, Cicirelli G and Distante A. An algorithm for real time eyedetection in face images[C]. Proeedings of International Conference on PatternRecognition,2004,3:278-281.
[70] Zhou Z and Geng X. Projection functions for eye detection[J]. Pattern Recogn,2004,37(5):1049-1056.
[71] Fu Y, Yan H, Li J and Xiang R. Robust facial features localization on rotation arbitrarymulti-view face in complex background[J]. Journal of Computers,2011,6(2):337-342.
[72] Ma Y, Ding X, Wang Z and Wang N. Robust precise eye location under probabilisticframework[C]. Proceedings of IEEE Conference on Automatic Face and GestureRecognition,2004:339-344.
[73] Campadelli P, Lanzarotti R and Lipori G.. Precise eye localization through ageneral-to-specific model definition[C]. Proceedings of IEEE Conference on BritishMachine Vision Conf,2006:187-197.
[74] Hamouz M, Kittlerand J, Kamarainen J K, Paalanen P, Kalviainen H and Matas J.Feature-based affine-invariant localization of faces[J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,2005,27(9):1490-1495.
[75] Yang M H, Ahuia N, Kriegman D. Detecting faces in images: a survey[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,24(1):34-58.
[76] Lee T S. Image representation using2D Gabor wavelets[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,1996,18(10):959-971.
[77] Asteriadis S, Nikolaidis N and Pitas I. Facial feature detection using distance vectorfields[J]. Pattern Recognition,2009,42(7):1388-1398.
[78] Liu S, Tian Y and Wan C. Gabor feature representation method based on blockstatistics and its application to facial expression recognition[C]. Proceedings of the8thWorld Congress on Intelligent Control and Automation. Ji’nan, China: IEEE,2010,6267-6271.
[79] Jesorsky O, Kirchberg K and Frischholz R. Robust face detection using the Hausdorffdistance[C]. Bigun, J. Smeraldi, F.(Eds.), Lecture Notes in Computer Science,2001:90–95.
[80] Zhou Z and Geng X. Projection functions for eye detection[J]. Pattern Recogn,2004,37(5):1049-1056.
[81] Fu Y, Yan H, Li J and Xiang R. Robust facial features localization on rotation arbitrarymulti-view face in complex background[J]. Journal of Computers,2011,6(2):337-342.
[82] Ma Y, Ding X, Wang Z and Wang N. Robust precise eye location under probabilisticframework[C]. Proceedings of IEEE Conference on Automatic Face and GestureRecognition,2004:339-344.
[83] Mase K. Recognition of facialexpression from optical flow[J]. IEICE Transactions,1991,74(10):3474-3484.
[84] Yacoob Y, Davis L. Recognizing human facial expressions form long imagesequencesusing optical flow[J]. IEEE Transactionson Pattern Analysis andMachineIntelligence,1994,16(6):636-642.
[85] Otsuka T, Ohya J. Recognizing multiple persons facial expressions using HMM basedon automatic extraction of significant frames from image sequence[C]. Proceedings ofthe InternationalConference on Image Processing, California, USA,1997:546-549.
[86] Donato G, Bartlett M S, Hager J C, et al. Classifying facial actions[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,1999,21(10):974-989.
[87] Zhang L G, Tjondronegoro, Dian W, Chandran, Vinod. Toward a more robust facialexpression recognition in occluded images using randomly sampled Gabor basedtemplates[C]. Proceedings of the IEEE International Conference on Multimedia andExposition,2011:1-6.
[88] Daugrnan J. Uneertainty relation for resolution in space, spatial frequeney andorientation optimized by two-dimensional visual cortieal filters[J]. Joumal of theOptieal Soeiety of Ameriea A,1985,(2):1160-1169.
[89] Campell F W, Robson J G. Applieation of Fourier analysis to the visibility of gratings[J]. Physiology,1968,(197):551-556.
[90] Van den Bergh F. An analysis of particle swarm optimizer[D]. Proceedings of the1998conference of Evolutionary computation,1998:67-73.
[91]杨耿煌，温渤婴．基于量子行为粒子群优化–人工神经网络的电能质量扰动识别
[J]．中国电机工程学报，2008，28(10)：123-129．
[2] Lyons M, Akamatsu S, Kamachi M, et al. Coding facial expressions with Gaborwavelets[C]. Proceeding of the Thrid IEEE International Conference on Automatic Faceand Gesture Recognition. Nara, Japan,1998:200-205.
[3] Wong S K and Ziarko W. On optimal decision rules in decision tables[J]. Bulletin ofPolish Academy of Science,1985,33:693696.
[4] Mikolajczyk K, Tuytelaars T, Schmid C, et al. A comparison of affine regiondetectors[J]. International Journal of Computer Vision,2005,65(1-2):43-72.
[5] Belhumeur P N, Hespanha J, Kriegman D J. Eigenfaces vs. fisherfaces: recognitionusing class specific linear projection[J]. IEEE Pattern Analysis and MachineIntelligence,1997,19(7):711-720.
[6] Huang R, Liu Q and Lu H. Solving the small sample size problem of LDA[C]. InProceeding of16th International Conference on Pattern Recognition (ICPR2002).Quebec, Canada: IEEE Computer Society,2002:29-32.
[7] Dai D Q and Pong C. Yuenv. Face recognition by regularized discriminant analysis[J].Systems, Man, and Cybernetics part B-cybernetics,2007,37(4):1080-1085.
[8] Mendes R, Kenney J. The fully informed particle swarm: Simpler, maybe better[J].IEEE Transaction on Evolutionary Computation,2004,8(3):204-210.
[9] Ye J and Li Q. A two-stage linear discriminant analysis via QR-decomposition[J].Pattern Analysis and Machine Intelligence,2005,27(6):929-941.
[10] Chen L F, Liao H M and Ko M T. A new LDA-based face recognition system whichcan solve the small sample size problem[J]. Pattern Recognition,2000,33(10):1713-1726.
[11] Liu J and Chen S. Discriminant common vecotors versus neighbourhood componentsanalysis and laplacianfaces: A comparative study in small sample size problem[J].Image and Vision Computing,2006,24(3):249-262.
[12] Wang X and Tang X. Random sampling for subspace face recognition[J]. InternationalJournal of Computer Vision,2006,70(1):94-104.
[13] Chawla N V and Bowyer K W. Random subspaces and subsampling for2-D facerecognition[C]. Computer Vision and Pattern Recognition (CVPR2005). San Diego,USA: IEEE Computer Society,2005:582-589.
[14] Wang X, Huang C and Fang X.2DPCA vs.2DLDA: face recognition usingtwo-dimensional method[C]. Artificial Intelligence and Computational Intelligence.Shanghai, China: Springer,2009:357-360.
[15] Y Jian, Zhang D, Yong X and Yang J. Two-dimensional discriminant transform forface recognition[J]. Pattern Recognition, July2005,38(7):1125-1129.
[16] Noushath S, Hemantha Kumar G and Shivakumara P.(2D)2LDA: An efficientapproach for face recognition[J]. Pattern Recognition,2006,39(7):1369-1400.
[17] Luo D, Ding C and Huang H. Symmetric two dimensional linear discriminantanalysis(2DLDA)[C]. Computer Vision and Pattern Recognition (CVPR2009). Miami, FL,USA: IEEE Computer Society,2009:2812-2819.
[18] Wang X, Huang C and Liu J. Gabor-2DLDA: face recognition using Gabor featuresand2D linear discriminant analysis[C]. Computation Technology and Automation.Changsha, China: IEEE Computer Society,2009:608-610.
[19] Wang J, Lu H, Plataniotis K N and Lu J. Gaussian kernel optimization for patternclassification[J]. Pattern Recognition, July2009,42(7):1237-1247.
[20] Wang J, Plataniotis K N, Lu J and Venetsanopoulos A N. Kernel QuadraticDiscriminant analysis for small sample size problem[J]. Pattern Recognition,2008,39(5):1528-1538.
[21] Lu J, Plataniotis K N and Venetsanopoulos A N. Kernel discriminant learning withapplication to face recognition[C]. Support Vector Machines: Theory and Applications.Berlin Heidelberg: Springer-Verlag,2005:275-296.
[22] Lu J, Plataniotis K N, Venetsanopoulos A N and Wang J. An efficient kerneldiscriminant analysis method[J]. Pattern Recognition,2005,38(10):1788-1790.
[23] Lu J, Plataniotis K N and Venetsanopoulos A N. Face recognition using kernel directdiscriminant analysis algorithms[J]. IEEE Transactions on Neural Networks, January2003,14(1):117-126.
[24] Wang J, Plataniotis K N, Lu J and Venetsanopoulos A N. On solving the facerecognition problem with one training sample per subject[J]. Pattern Recognition,2006,39(9):1746-1762.
[25] Liu J, Chen S and Zhou Z. Single image subspace for face Recognition[C]. Analysisand modeling of faces and gestures. Rio de Janeiro, Brazil: Springer-Verlag Berlin,Heidelberg,2007:205-219.
[26] Chen S, Liu J and Zhou Z. Making FLDA application to face recognition with onesample per person[J]. Pattern Recognition,2004,37(7):1553-1555.
[27] Zheng W, Zhao L and Zou C. An efficient algorithm to solve the small sample sizeproblem for LDA[J]. Pattern Recognition, May2004,37(5):1077-1079.
[28] Zhou J, Lu C, Zhang C and Li Y. A survey of automatic human face recognition[J].Acta Electronica Sinica,2000,28(4):102-106.
[29] Tao D, Tang X, Li X and Rui Y. Kernel direct biased discriminant analysis: a newcontent-based image retrieval relevance feedback algorithm[J]. IEEE Transactions onMultimedia,2006,8:716-727.
[30] Philbin J, Chum O, Isard M, Sivic J and Zisserman A. Object retrieval with largevocabularies and fast spatial matching[C]. Proceedings of the IEEE Conference onComputer Vision and Pattern Recognition,2007:1-8.
[31] Schaffalitzky F and Zisserman A. Multi-view matching for unordered image sets[M].Proceedings of the European Conference on Computer Vision,2002:414-431.
[32] Tuytelaars T and Van Gool L. Matching widely separated views based on affineinvariant regions[J]. International Journal of Computer Vision,2004,59(1):61-85.
[33] Ferrari V, Tuytelaars T and Van Gool L. Simultaneous object recognition andsegmentation by image exploration[M]. Proceedings of the European Conference onComputer Vision,2004:40-54.
[34] Lazebnik S, Schmid C and Ponce J. Sparse texture representation usingaffine-invariant neighborhoods[C]. Proceedings of the IEEE Conference on ComputerVision and Pattern Recognition,2003:319-324.
[35] Fergus R, Perona P and Zisserman A. Object class recognition by unsupervisedscale-invariant learning[C]. Proceedings of the IEEE Conference on Computer Visionand Pattern Recognition,2003,2:264-271
[36] Brown M and Lowe D. Recognising panoramas[C]. Proceedings of InternationalConference on Computer Vision,2003,3:1218-1227.
[37] Ferrari V, Tuytelaars T and Van Gool L. Wide-baseline multiple-viewcorrespondences[C]. Proceedings of the IEEE Conference on Computer Vision andPattern Recognition,2003,1:718-725.
[38]王永明，王贵锦.图像局部不变性特征与描述[B]，北京：国防工业出版，2010.
[39] Moravec H P. Towards automatic visual obstacle avoidance[C]. Proceedings ofInternational Conference on Artificial Intelligence,1977:584.
[40] Harris C and Stephens M. A combined corner and edge detector[C]. Alvey VisionConference,1988:147-151.
[41] Smith S M and Brady J M. SUSAN-a new approach to low level image[J].Processing, International Journal of Computer Vision,1997,23(1):45-78.
[42] Matas J, Chum O, Urban M and Pajdla T. Robust wide-baseline stereo frommaximally stable extremal regions[J]. Proceedings of British Machine VisionConference,2004,22(10):761-767.
[43] Lowe D G.. Distinctive image features from scale-invariant keypoints[J]. InternationalJournal of Computer Vision,2004,60(2):91-110.
[44] Mokhtarian F and Suomela R. Robust image corner detection through curvaturescale-space[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(12):1376–1381.
[45] Mikolajczyk K and Schmid C. Scale and affine invariant interest point detectors[J].International Journal on Computer Vision,2004,60(1):63-86.
[46] Tuytelaars T and Van Gool L. Matching widely separated views based on affineinvariant regions[J]. International Journal of Computer Vision,2004,59(1):61-85.
[47] Matas J, Chum O, Urban M and Pajdla T. Robust wide-baseline stereo frommaximally stable extremal regions[J]. Proceedings of British Machine VisionConference,2004,22(10):761-767.
[48] Kadir T, Zisserman A and Brady M. An affine invariant salient region detector[M].Proceedings of the European Conference on Computer Vision,2004,3021:228-241.
[49] Mikolajczyk K and Schmid C. A performance evaluation of local descriptors[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2005,27(10):1615-1630.
[50] Ma L and Khorasani K. Facial expression recognition using constructive feedforwardNeural Networks[J]. IEEE Transactions on Systems Man and Cybernetics,2004,34(3):1588-1595.
[51] Gueorguieva N, Georgiev G and Valova I. Facial expression recognition usingfeedforward neural networks[C]. Proceedings of the International Conference onArtificial Intelligence, Las Vegas, N V, USA,2003:285-291.
[52] Soyel H and Demirel H. Facial expression recognition using3D facial featuredistances[C]. Proceedings of the4th International Conference on Image Analysis andRecognition, Montreal, Canada,2007:831-838.
[53]周书仁，梁昔明，杨秋芬.类间学习神经网络的人脸表情识别[J].计算机应用研究，2008，25(7)：2219-2222.
[54] Sebe N, Cohen I, Garg A, et al. Emotion recognition using a Cauchy naive BayesClassifier[C]. Proceedings of International Conference on Pattern Recognition,Québec City, Canada,2002,(1):17-20.
[55] Cohen I, Sebe N, Garg A, et al. Facial expression recognition from video sequences:Temporal and staticmodeling[J]. Computer Vision and Image Understanding,2003,91(1-2):160-187.
[56] Cohen I, Sebe N, Cozman F G, et al. Learning bayesian network classifiers for facialexpression recognition with both labeled and unlabeled data[C]. Proceedings ofInternational Conference on Computer Vision and Pattern Recognition, Madison,Wisconsin, USA,2003,(1):595-604.
[57] Zhang Y and Ji Q. Active and dynamic information fusion for facial expressionunderstanding from image sequences[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,2005,27(5):699-714.
[58]朴林植，赵杰煜.基于统计形状分析的人脸基本表情分析[J].宁波大学学报(理工版),2009，22(2)：196-210.
[59] Littlewort G, Bartlett M S, Fasel I, et al. Analysis of machine learning methods forreal-time recognition of facial expressions from video[C]. Machine PerceptionLaboratory Institute for Neural Computation, University of California, San Diego, CA,USA,2004:27.
[60] Xu Q Z, Zhang P Z, Yang L X, et al. A facial expression recognition approach basedon novel support vector machine tree[C]. Proceedings of the4th InternationalSymposium on Neural Networks, Nanjing, China,2007:374-381.
[61] Bartlett M S, Littlewort G, Frank M, et al. Recognizing facial expression: machinelearning and application to spontaneous behavior[C]. Proceedings of IEEE ComputerSociety Conference on Computer Vision and Pattern Recognition, San Diego, CA,USA,2005,2:568-573.
[62]孙雯玉.人脸表情识别算法研究[D].北京交通大学硕士学位论文,2007.
[63] Muller S, Wallhoff F, Hulsken F, et al. Facial expression recognition using pseudo3-Dhidden Markov models[C]. Proceedings of International Conference on PatternRecognition, Québec City, Canada,2002,2:32~35.
[64] Yeasin M, Bullot B and Sharma R. From facial expression to level of interest: aspatio-temporal approach[C]. Proceedings of International Conference on ComputerVision and Pattern Recognition, Washington, DC, USA,2004,2:922-927.
[65] Shan S, Chang Y and Gao W. Curse of mis-alignment in face recognition: problem anda novel mis-alignment learning solution[C]. Proeedings of International Conference onAutomatic Face and Gesture Recognition,2004,314–320.
[66] Wang P, Green M, Ji Q and Waymanm J. Automatic eye detection and its validation[C].Proeedings of International Conference on Computer Vision and Pattern Recognition,2005,164–164.
[67] Rodriguez Y, Cardinaux F, Bengio S and Marithoz J. Measuring the performance offace localization systems[J]. Image Vision Comput,2006,24(8):882-893.
[68] Haro A, Flickner M and Essa I. Detecting and tracking eyes by using theirphysiological properties, dynamics, and appearance[C]. Proeedings of InternationalConference on Computer Vision and Pattern Recognition,2000,163-168.
[69] D’Orazio T, Leo M, Cicirelli G and Distante A. An algorithm for real time eyedetection in face images[C]. Proeedings of International Conference on PatternRecognition,2004,3:278-281.
[70] Zhou Z and Geng X. Projection functions for eye detection[J]. Pattern Recogn,2004,37(5):1049-1056.
[71] Fu Y, Yan H, Li J and Xiang R. Robust facial features localization on rotation arbitrarymulti-view face in complex background[J]. Journal of Computers,2011,6(2):337-342.
[72] Ma Y, Ding X, Wang Z and Wang N. Robust precise eye location under probabilisticframework[C]. Proceedings of IEEE Conference on Automatic Face and GestureRecognition,2004:339-344.
[73] Campadelli P, Lanzarotti R and Lipori G.. Precise eye localization through ageneral-to-specific model definition[C]. Proceedings of IEEE Conference on BritishMachine Vision Conf,2006:187-197.
[74] Hamouz M, Kittlerand J, Kamarainen J K, Paalanen P, Kalviainen H and Matas J.Feature-based affine-invariant localization of faces[J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,2005,27(9):1490-1495.
[75] Yang M H, Ahuia N, Kriegman D. Detecting faces in images: a survey[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,24(1):34-58.
[76] Lee T S. Image representation using2D Gabor wavelets[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,1996,18(10):959-971.
[77] Asteriadis S, Nikolaidis N and Pitas I. Facial feature detection using distance vectorfields[J]. Pattern Recognition,2009,42(7):1388-1398.
[78] Liu S, Tian Y and Wan C. Gabor feature representation method based on blockstatistics and its application to facial expression recognition[C]. Proceedings of the8thWorld Congress on Intelligent Control and Automation. Ji’nan, China: IEEE,2010,6267-6271.
[79] Jesorsky O, Kirchberg K and Frischholz R. Robust face detection using the Hausdorffdistance[C]. Bigun, J. Smeraldi, F.(Eds.), Lecture Notes in Computer Science,2001:90–95.
[80] Zhou Z and Geng X. Projection functions for eye detection[J]. Pattern Recogn,2004,37(5):1049-1056.
[81] Fu Y, Yan H, Li J and Xiang R. Robust facial features localization on rotation arbitrarymulti-view face in complex background[J]. Journal of Computers,2011,6(2):337-342.
[82] Ma Y, Ding X, Wang Z and Wang N. Robust precise eye location under probabilisticframework[C]. Proceedings of IEEE Conference on Automatic Face and GestureRecognition,2004:339-344.
[83] Mase K. Recognition of facialexpression from optical flow[J]. IEICE Transactions,1991,74(10):3474-3484.
[84] Yacoob Y, Davis L. Recognizing human facial expressions form long imagesequencesusing optical flow[J]. IEEE Transactionson Pattern Analysis andMachineIntelligence,1994,16(6):636-642.
[85] Otsuka T, Ohya J. Recognizing multiple persons facial expressions using HMM basedon automatic extraction of significant frames from image sequence[C]. Proceedings ofthe InternationalConference on Image Processing, California, USA,1997:546-549.
[86] Donato G, Bartlett M S, Hager J C, et al. Classifying facial actions[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,1999,21(10):974-989.
[87] Zhang L G, Tjondronegoro, Dian W, Chandran, Vinod. Toward a more robust facialexpression recognition in occluded images using randomly sampled Gabor basedtemplates[C]. Proceedings of the IEEE International Conference on Multimedia andExposition,2011:1-6.
[88] Daugrnan J. Uneertainty relation for resolution in space, spatial frequeney andorientation optimized by two-dimensional visual cortieal filters[J]. Joumal of theOptieal Soeiety of Ameriea A,1985,(2):1160-1169.
[89] Campell F W, Robson J G. Applieation of Fourier analysis to the visibility of gratings[J]. Physiology,1968,(197):551-556.
[90] Van den Bergh F. An analysis of particle swarm optimizer[D]. Proceedings of the1998conference of Evolutionary computation,1998:67-73.
[91]杨耿煌，温渤婴．基于量子行为粒子群优化–人工神经网络的电能质量扰动识别
[J]．中国电机工程学报，2008，28(10)：123-129．