基于图像的人体检测跟踪和人脸识别的研究
|
摘要
基于图像的人体检测跟踪和人脸识别问题是当今计算机视觉和模式识别领域的热点研究问题，它在图像处理、智能监控、智能汽车系统等领域有着广泛的应用前景。本文针对图像中运动人体检测跟踪和人脸识别中的一些关键问题进行研究，并取得了一定的进展，具体有如下四方面：
在人体检测方面，提出了基于二次连通域处理的人体检测方法，这种方法首先提取运动目标并作数学形态学处理，然后采用四方向连接方法去除空洞，第一次连接断开区域，接着采用三次扫描标记法第二次连接断开区域，随后提取HOG特征，利用Adaboost训练强分类器进而识别运动前景是否是人体。
在人体跟踪方面，提出了基于改进Meanshift的人体跟踪方法，这种改进方法是通过判别跟踪区域是背景区域或前景区域来设置权重系数，同时将跟踪区域精确到运动人体，再结合卡尔曼滤波器来预测运动人体下一步的方向，由于降低了背景区域的计算量，所以跟踪效果更好些。
在光照预处理方面，提出了改进高斯差分滤波的光照预处理方法，通过改变高斯差分滤波器中水平方向和垂直方向的参数，进而将原来的圆形滤波器扩展为椭圆形滤波器，更加适合人脸的面部特征，对于光照不均匀情况适应性较好，结合人脸识别算法获得了较高的人脸识别率和较低的人脸误识率。
在人脸识别方面，提出了基于Adaboost的双向2DLDA融合的人脸识别方法，即2DLDA和E2DLDA的融合。2DLDA主要利用图像垂直方向上的判别信息，E2DLDA主要利用图像水平方向上的判别信息，然后利用Adaboost融合这两个方向的判别信息，实验表明整体的识别算法具有更好的识别性能。
Computer vision usually means the process of obtaining visual information andpost-processing by computer control and sensor equipment from the surrounding environment,including expression, compression, analysis, processing, storage, and so on, so as to realizethe human biological vision of the function "look". In recent years, the rapid development ofcomputer technology, photovoltaic technology and automation technology, with theemergence of the computer vision system and widely used in the continuous improvement ofscience, industrial, medical and military fields, computer vision technology has become a hotspot of research. Research on computer vision systems has great value and significance bothin theoretical research and practical application research.
Human detection tracking and face recognition is the hot research field in computer visionand pattern recognition, it has a broad application prospects in the field of graphics, imageprocessing, intelligent monitoring, smart car system and so on. This paper focuses on somekey issues of human detection tracking and face recognition in image. The details are asfollows:
In aspect of moving human detection, human recognition method is widely used in allkinds of videos. But recognition accuracy of these methods is changed negatively because ofcomplexity of background, e.g. void and noise. So a moving body recognition method withbackground robustness is proposed to solve this problem. This method uses tripling temporaldifference method to recognize moving object and dilate the computed binary image withmathematical morphology. And then it uses quadruple directions connection method toconnect disconnection areas. Finally, train the classifier by HOG feature to identify whetherthe moving target is human or not. This method not only needs no restraint of background, butalso shows less computation and higher accurate. Experiment results show its positivecomputational complexity, accuracy and robustness.
In aspect of moving human tracking, classic meanshift algorithm is widely used in patternrecognition, but its accuracy decrease large when there is perturbation in background. So tosolve this problem, an improved meanshift method to recognize and track moving body isproposed. This method resets weighting coefficient automatically by detecting tracking area isforeground or not. To compare with exist method, the new method decreases computationamount of background, changed tracking area and predict to moving body. So tracking effectis better than others. Experiment result validates accuracy, effectiveness and robustness.
In aspect of illumination pretreatment, illumination pretreatment is an important researchdomain of face recognition, and it has a very important influence to the face recognition, to solve the problem of abnormal light, an improved DOG algorithm to make illuminationpretreatment is proposed. To combine with face reorganization method, compare and analyzelots of characters with experiments of major illumination pretreatment algorithms. Fromexperiment results, the improved DOG algorithm is better than any other ones when there arechanges in illumination and expression of images.
In aspect of face recognition, a challenge for face recognition is variation, such as due tolighting or facial expression differences. To solve this problem, this method fusesbidirectional two-dimensional linear discriminant analysis (2DLDA) feature by adaboosttechnique and propose a novel recognition method called AB2DLDA. It can perform wellwith small number of samples.This method uses adaboost to design a classifier and fuses2DLDA and E2DLDA method to improve recognition performance. Experimental resultsshow that the method has good recognition accuracy and robustness.
The research based on the moving body detection tracking and face recognition in image isan important subject in a wide range field of computer vision, there are still many difficultpoints which need further research, may this article could give the researchers someinspiration and help.
在人体检测方面，提出了基于二次连通域处理的人体检测方法，这种方法首先提取运动目标并作数学形态学处理，然后采用四方向连接方法去除空洞，第一次连接断开区域，接着采用三次扫描标记法第二次连接断开区域，随后提取HOG特征，利用Adaboost训练强分类器进而识别运动前景是否是人体。
在人体跟踪方面，提出了基于改进Meanshift的人体跟踪方法，这种改进方法是通过判别跟踪区域是背景区域或前景区域来设置权重系数，同时将跟踪区域精确到运动人体，再结合卡尔曼滤波器来预测运动人体下一步的方向，由于降低了背景区域的计算量，所以跟踪效果更好些。
在光照预处理方面，提出了改进高斯差分滤波的光照预处理方法，通过改变高斯差分滤波器中水平方向和垂直方向的参数，进而将原来的圆形滤波器扩展为椭圆形滤波器，更加适合人脸的面部特征，对于光照不均匀情况适应性较好，结合人脸识别算法获得了较高的人脸识别率和较低的人脸误识率。
在人脸识别方面，提出了基于Adaboost的双向2DLDA融合的人脸识别方法，即2DLDA和E2DLDA的融合。2DLDA主要利用图像垂直方向上的判别信息，E2DLDA主要利用图像水平方向上的判别信息，然后利用Adaboost融合这两个方向的判别信息，实验表明整体的识别算法具有更好的识别性能。
Computer vision usually means the process of obtaining visual information andpost-processing by computer control and sensor equipment from the surrounding environment,including expression, compression, analysis, processing, storage, and so on, so as to realizethe human biological vision of the function "look". In recent years, the rapid development ofcomputer technology, photovoltaic technology and automation technology, with theemergence of the computer vision system and widely used in the continuous improvement ofscience, industrial, medical and military fields, computer vision technology has become a hotspot of research. Research on computer vision systems has great value and significance bothin theoretical research and practical application research.
Human detection tracking and face recognition is the hot research field in computer visionand pattern recognition, it has a broad application prospects in the field of graphics, imageprocessing, intelligent monitoring, smart car system and so on. This paper focuses on somekey issues of human detection tracking and face recognition in image. The details are asfollows:
In aspect of moving human detection, human recognition method is widely used in allkinds of videos. But recognition accuracy of these methods is changed negatively because ofcomplexity of background, e.g. void and noise. So a moving body recognition method withbackground robustness is proposed to solve this problem. This method uses tripling temporaldifference method to recognize moving object and dilate the computed binary image withmathematical morphology. And then it uses quadruple directions connection method toconnect disconnection areas. Finally, train the classifier by HOG feature to identify whetherthe moving target is human or not. This method not only needs no restraint of background, butalso shows less computation and higher accurate. Experiment results show its positivecomputational complexity, accuracy and robustness.
In aspect of moving human tracking, classic meanshift algorithm is widely used in patternrecognition, but its accuracy decrease large when there is perturbation in background. So tosolve this problem, an improved meanshift method to recognize and track moving body isproposed. This method resets weighting coefficient automatically by detecting tracking area isforeground or not. To compare with exist method, the new method decreases computationamount of background, changed tracking area and predict to moving body. So tracking effectis better than others. Experiment result validates accuracy, effectiveness and robustness.
In aspect of illumination pretreatment, illumination pretreatment is an important researchdomain of face recognition, and it has a very important influence to the face recognition, to solve the problem of abnormal light, an improved DOG algorithm to make illuminationpretreatment is proposed. To combine with face reorganization method, compare and analyzelots of characters with experiments of major illumination pretreatment algorithms. Fromexperiment results, the improved DOG algorithm is better than any other ones when there arechanges in illumination and expression of images.
In aspect of face recognition, a challenge for face recognition is variation, such as due tolighting or facial expression differences. To solve this problem, this method fusesbidirectional two-dimensional linear discriminant analysis (2DLDA) feature by adaboosttechnique and propose a novel recognition method called AB2DLDA. It can perform wellwith small number of samples.This method uses adaboost to design a classifier and fuses2DLDA and E2DLDA method to improve recognition performance. Experimental resultsshow that the method has good recognition accuracy and robustness.
The research based on the moving body detection tracking and face recognition in image isan important subject in a wide range field of computer vision, there are still many difficultpoints which need further research, may this article could give the researchers someinspiration and help.
引文
[1] David A.Forsy. Computer Vision: A Modern Approach Second Edition[M].北京：电子工业出版社,2012.
[2] Marr D.姚国正等译.视觉计算理论[M].北京：科学出版社,1988.
[3]崔之枯，江春，陈丽鑫.数字视频处理[M].北京:电子工业出版社,1998.
[4]刘小军.人脸识别技术研究[D].中国科学院.2001.
[5] Prange J. Detecting, recognizing and understanding video events in surveillancevideo[J].Advanced Video and Signal Based Surveillance,2003,4.
[6]王亮,胡卫明,谭铁牛．人运动的视觉分析综述[J]．计算机学报，2002,25(3):1-16.
[7]边肇祺,张学工.《模式识别》第二版[M].北京：清华大学出版社,2001：223-227.
[8] D.M.Gavrila. The Visual Analysis of Human Movement: A Survey. Computer Vision andImage Understanding.1999,73(1):82-98.
[9] Ismail Haritaoglu,David Harwood,Larry S.Davis.W4:Real-Time Surveillance of Peopleand Their Activities[J].IEEE Transactions on Pattern Analysis and MachineIntelligence.2000,22(8):809-830.
[10]Matsuyama T, Ukita N. Real-time multitarget tracking by a cooperative distributedvision system[J].Proceedings of the IEEE,2002,90:1136-1150.
[11]Shu C, Hampapur A, Lu M. IBM smart surveillance system (S3):an open and extensiblearchitecture for smart video surveillance[C]. In IEEE International Conference onAdvanced Video and Signal based Surveillance,2005,318-323.
[12]Harmon L D. The recognition of faces[J].Sci.1973,229:71-82.
[13]T Kanade. Picture processing system by computer and recognition of human faces[D].Kyoto University,1973.
[14]Baron RJ. Mechanisms of human facial recognition [J]. Man-machine Studies.1981,15:137-178.
[15]Alex Pentland, Baback Moghaddam, Thad starner. View-Based and ModularEigen-spaces for Face Recognition[C]. MIT Media Lab perceptual Computing SectionTech IEEE Conference on Computer Vision&Pattern Recognition,1994.
[16]P N Bellhumer, J Hespanha, D Kriegman. Eigenfaces vs.fisherfaces: Recognition usingclass specific linear projection [J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1997,17(7):711-720.
[17]W.Zhao,R.Chellappa. SFS Based View Synthesis for Robust Face Recognition[C], InProceeding of the4th International Conference on Face and Gesture Recognition,2000,285-292.
[18]C Liu, H Wechsler. Gabor Feature Based Classification Using the Enhanced Fisher LinearDiscriminant Model for Face Recognition [J].IEEE Trans.Image Processing.2002,11(4):467-476.
[19]H Yu, J Yang. A Direct LDA Algorithm for High-dimensional Data–with Application toFace Recognition[J]. Pattern Recognition.2001,34:2067-2070.
[20]刘青山.人脸跟踪与识别的研究[D].中科院自动化所.2003.
[21]Ming Li，Baozong Yuan．2D-LDA:A Statistical Linear Discriminant Analysis for ImageMatrix[J]．Pattern Recognition Letters．2005,26(5):527-532.
[22]B.Moghaddam, A.Pentland. Probabilistic Visual Learning for Object representation [J].IEEE Trans.On PAMI.1997,20(7):696-710.
[23]Baback Moghaddam, Tony Jebara, Alex Pentland. Bayesian Face Recognition [J]. PatternRecognition.2000,33:1771-1782.
[24]J Buhmann, M Lades, von der Malsburg. Size and distortion invariant object recognitionby hierarchical graph matching[C]. In Proceedings of IEEE Intl. Joint Conference onNeural Networks. Santiago,1990,411-416.
[25]A Tefas, C Kotropoulos, I Pitas. Using Support Vector Machines to Enhance thePerformance of Elastic Graph Matching for Frontal Face Authentication [J]. IEEETransactions On Pattern Analysis And Machine Intelligence.2001,23(7):735-746.
[26]丁嵘,苏光大,林行刚.使用关键点信息改进弹性匹配人脸识别算法[J].电子学报.2002,30(9):1292-1294.
[27]Hochul Shin, Seong Dae Kim, Hae Chul Choi. Generalized elastic graph matching forface recognition [J]. Pattern Recognition Letters.2007,28(9):1077-1082.
[28]P.Penev, J.Atick. Local Feature Analysis: A General Statistical Theory for ObjectRepresentation [J]. Computation in Neural Systems.1996,7:477-500.
[29]A Lanitis, C J Taylor, T F Cootes. An automatic face identification system using flexibleappearance models [C]. In British Machine Vision Conference.1994,65-74.
[30]A Lanitis, C Taylor, T Cootes. A Unified Approach to Coding and Interpreting FaceImages [C]. In Proceedings of Int'l Conference Computer Vision.1995,368-373.
[31]T F Cootes, G J Edwards, C J Taylor. Active Appearance Models [C], In Proceedings ofEuropean Conference Computer Vision.1998,484-498.
[32]G Edwards, T Cootes, C Taylor. Advances in Active Appearance Models, In Proceedingsof Int'l Conference Computer Vision.1999,137-142.
[33]M B Stegmann. Active Appearance Models, Theory, Extension and Cases, Master thesis[D]. Technical University of Denmark.2000.
[34]S C Yan, C Liu, S Z Li, H J Zhang, H Shum,Q S Cheng. Face Alignment UsingTexture-Constrained Active Shape Models [J]. Image and Vision Computing.2003,21(1):69-75.
[35]Tong Wang, Haizhou Ai, Gaofeng Huang. A Two-Stage Approach to Automatic FaceAlignment [C]. In Proceedings of SPIE-The International Society for Optical Engineering.2003,558-563.
[36]Yi Zhou, Lie Gu, Hongjiang Zhang. Bayesian Tangent Shape Model: Estimating Shapeand Pose Parameters via Bayesian Inference [C]. In Proceeding of The IEEE Conferenceon Computer Vision and pattern Recognition. USA: Wisconsin,2003,16-22.
[37]Shiguang Shan, Wen Gao, Wei Wang, Debin Zhao, Baocai Yin. Enhanced Active ShapeModels with Global Texture Constraints for Face Image Analysis[C]. FourteenthInternational Symposium On Methodologies For Intelligent Systems. Japan: MaebashiCity,2003,593-597.
[38]Fatih Kahraman, Mikkel B Stegmann. Towards Illumination-invariant Localization ofFaces using Active Appearance Models [C]. The7th Nordic Signal ProcessingSymposium. Iceland: Reykjavik,2006.
[39]P J Philips, P J Grother, R J Micheals, D M Blackburn, E Tabassi, J M Bone. FaceRecognition Vendor Test2002: Evaluation report. Technical Report. National Institute ofStandards and Technology.2003.
[40]G Guo, S Z Li, K Chan. Face Recognition by Support Vector Machines [C]. InProceedings of the4th International Conference on Auto. Face and Gesture Recognition.France: Grenoble,2000,196-201.
[41]P J Phillips. Support vector machines applied to face recognition [C]. In Advances inNeural Information Processing Systems.1998,803.
[42]Volker Blanz, Thomas Vetter. A Morphable Model For the Synthesis of3D Faces [J].SIG’ GRAPH’,1999.
[43]Sami Romdhani, Volker Blanz, Thomas Vetter. Face Identification by Fitting a3DMorphable Model Using Linear Shape and Texture Error Functions [C]. In Proceedingsof the7th European Conference on Computer Vision,2002,3-19.
[44]Y Freund, R E Schapire. A decision-theoretic generalization of on-line learning and anapplication to boosting [J]. Journal of Computer and System Sciences,,1997,55(1):119-139.
[45]R Meir, G Raetsch. An introduction to Boosting and Leveraging [J]. Advanced Lectureson Machine Learning,2003:118-183.
[46]J Lu, KN Plataniotics, A N Venetsanopoulos. Ensemble-based discriminant learning withboosting for face recognition [J]. IEEE transactions on neural networks,2006,17(1):166-178.
[47]张文超,山世光,张洪明,陈杰,陈熙霖,高文.基于局部Gabor变化直方图序列的人脸描述与识别[J].软件学报,2006,17(12):2508-2517.
[48]Xiangsheng Huang, Stan Z Li, Wang Yangsheng. Jensen-shannon boosting for objectrecognition [C]. International conference on computer vision and patternrecognition,2005.
[49]A S Georghiades, D J Kriegman, P N Belhumeur. Illumination Cones For RecognitionUnder Variable Lighting: Faces. In Proceedings of IEEE CVPR,1998,52-58.
[50]A Shashua, T Riklin Raviv. The Quotient Image: Class-Based Re-Rendering AndRecognition With Varying Illuminations [J]. IEEE Trans. on PAMI,2001,23(2):129-139.
[51]R Basri, D Jacobs. Lambertian Reflectance and Linear Subspaces [C],ICCV2001.Beckman Institute,2001,383-390.
[52]张志伟.可变光照条件下的人脸识别技术研究[D].河北工业大学.2007
[53]A R Martinez, R Benavente. The AR face database. Computer VisionCenter(CVC)Technical Report,1998.
[54]王先基.基于统计学习的自动人脸识别算法研究[D].中国科学技术大学.2007.
[55]卢官明,毕厚杰.基于数学形态学的图像序列分割[J].南京邮电学院学报,1997,17(2):54-57.
[56]韩军,熊琼,孙文彦等.自动分割及跟踪视频运动对象的一种实现方法[J].中国图像图形学报,2001,(8):732-738.
[57]Laurenz Wiskott. Segmentation from Motion:Combining Gabor-and Mallat-Wavelets toOvercome the Aperture and Correspondence Problems [J]. PatternRecognition,1999:1751-1766.
[58]黄友珍,黄艺,余兆明.基于修正分水岭算法和时域跟踪的视频自动分割[J].数字视频,2000:5-8.
[59]Gerald Kuhne, Stephan Richter, Markus Beier. Motion-Based Segmentation andContour-Based Classification of Video Objects [J]. ACM Multimedia,2001,6(9):254-273.
[60]黄波,杨勇,王桥等.一种用于视频分割德快速运动估计方法[J].电路与系统学报,2001,6(1):69-71.
[61]Y.Kim, H.Joon. Model-based Discontinuity Evaluation in the DCT Domain [J]. SignalProcessing,2001,81(4):871-882.
[62]詹劲峰,戚飞虎,王海龙.基于时空马尔可夫随机场德运动目标分割技术[J].通信学报,2000,21(11):63-83.
[63]Kuo Chung-Ming, Hsieh Chaur-Heh, Huang Yong-Ren. A New Mesh Based onTemporal-Spatial Segmentation for Image Sequence [C]. Computer Software andApplications Conference,2000:196-222.
[64]吴枫,高鹏,高文.基于网格模型的运动估计技术[J].电子学报,2000,28(5):47-51.
[65]夏利民,古士文.基于形变模型的图像分割和非刚性目标的跟踪[J].模式识别与人工智能,2000,13(1):64-68.
[66]Harper H.s, Dinggang shen. An Affine-Invariant Active Contour Model for Medel-BasedSegmentation [J]. Image and Vision Computing,1998,(16):135-146.
[67]Fan Jianping,Yu,Jun Gen Fujita. Spatio-Temporal Segmentation for Compact VideoRepresentation [J]. Signal Processing:Image Communication,2001,(16):553-566.
[68]施可为,傅锡天,蔡安妮等.含人脸的前景再活动图像序列中的分割[J].北京邮电大学学报,2000,23(1):66-70.
[69]张颖,张兆阳.基于二次空间变换的快速时空分割[J].中国图像图形学报,2000,5(9):744-749.
[70]Laurenz Wiskott. Segmentation from Motion:Combining Gabor-and Mallat-Wavelets toOvercome the Aperture and Correspondence Problems [J]. PatternRecognition,1999:1751-1766.
[71]Daniel Gatica-Perez, Sun,Ming-Ting Chuang Gu. Multlview Extensive PartitionOperators for Semantic Video Object Extraction [J]. IEEE Transactions on Circuits andSystems for Video Technology,2002,11(7):788-800.
[72]俞刚毅,黄艺.用于视频对象平面生成的运动对象自动分割[J].电子与信息学报,2001,23(7):657-662.
[73]N Friedman, S Russell. Image Segmentation In Video Sequences:A ProbabilisticApproach Uncertainty [J]. Artificial Intelligence.1997.
[74]A Elgammal, D Harwood, L Davis. Non-Parametric Model for Background Subtraction
[C]. In Proceedings IEEE Frame Rate Workshop.1999.
[75]Ismail Haritaoglu, David Harwood, Larry S.Davis. W4:Real-Time Surveillance of Peopleand Their Activities [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):809-830.
[76]ZENG Chun, LI Xiao-hua, ZHOU Ji-liu. Pedestrian detection based on HOG of ROI [J].Computer Engineering,2009,35(24):182-184.
[77]Lin Rui，Du Zhijiang，Sun Lining. Moving Object Tracking based on Mobile RobotVision [C], In Proceedings of International Conference on Mechatronics and Automation.2009.
[78]栾大勇.实时视频监控系统中的运动人体检测算法研究[D].吉林大学.2009.
[79]L.Wixson. Detecting Salient Motion by Accumulating Directionally-Consistent Flow[J].IEEE transaction on Pattern Analysis and Machine Intelligence,2000,22(8).:774-781.
[80]A.Shio, J Sklansky. Segmentation of people in motion [C]. In Proceeding IEEEWorkshop on Visual Motion,1991,325-332.
[81]Nuno Vasconcelos, Andrew Lippman. Empirical Bayesian Motion Segmentation [J].IEEE transactions on Pattern Analysis and Machine Intelligence,2001,23(2):217-221.
[82]WANG Xiaoyan，ZHANG Yanzhu，DONG Huiying，LI Yuan，LI Xiaojuan. AlgorithmResearch on Three-Frame Difference for Detection of Moving Target [J]. Journal ofShenyang Ligong University,2011,30(6):82-85.
[83]Navneet Dalal, Bill Triggs. Histograms of Oriented Gradients for Human Detection[C].In Proceedings of CVPR2005.
[84]Q Zhu, S Avidan, M C Yeh, K T Cheng. Fast Human Detection Using a Cascadeof Histogram of Oriented Gradients[C].In Proceedings of IEEE International Conference onComputer Vision and Pattern Recognition.2006.
[85]Bo Wu, Ram Nevatia. Detection of Multiple, Partially Occluded Humans in a SingleImage by Bayesian Combination of Edgelet Part Detectors[C]. In Proceedings ofICCV.2005.
[86]P Sabzmeydani, G Mori. Detecting pedestrians by learning shapelet features[C]. InProceedings of CVPR.2007.
[87]倪洪印.基于视频序列的人体检测和跟踪技术的研究[D].吉林大学.2010.
[88]刘海松.基于广义互补编码的目标识别处理器[J].中国激光，1999,26(3):1031-1035.
[89]Munchurl Kim, Jeon J G, Kwak J.S. Moving Object Segmentation in Video Sequences byUser Interaction and Automatic Object Tracking [J]. Image and Vision Computing,2001,19(5):245-258.
[90]Candemir Toklu, A Murat Tekalp, Tanju Erdem. Semi-Automatic Video Segmentation inthe Presence of Occlusion [J]. IEEE Transactions on Circuits and Systems for VideoTechnology,2002,10(4):624-629.
[91]Christopher Rasmussen, Gregory D.Hager. Probabilistic Data Association Method forTracking Complex Visual Objects [J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,2000,23(6):560-576.
[92]Holger Eggers, Fabrice Moscheni, Poberto Castagno. Robust Object Tracking Based onSpatial Characterization of Objects by Additive Invariant [J]. IEEE Transactions onCircuits and Systems for Video Technology,1999,9(9):844-848.
[93]Francois Bremond, Monique Thonnat. Tracking Multiple Non-Rigid Objects in VideoSequences [J]. IEEE Transactions on Circuits and Systems for VideoTechnology,1998,8(5):585-591.
[94]Zhang Jun, Gao Jianbo, Liu Wei-song. Image Sequence Segmentation Using3-DStructure Tensor and Curve Evolution [J]. IEEE Transactions on Circuits and Systems forVideo Technology,2001,11(5):629-641.
[95]Gerald Kuhne, Stephan Richter, Markus Beier. Motion-Based Segmentation andContour-Based Classification of Video Objects [J]. ACM Multimedia,2001,6(9):254-273.
[96]施建良,余松煜,傅立言.基于最短描述长度的序列图像运动分割[J].计算机学报,2001,22(8):809-815.
[97]Donna J.Williams, Mubarak Shah. A Fast Algorithm for Active Contours and CurvatureEstimation [J]. Image Understanding,1992,1(55):14-26.
[98]郭春,罗鹏飞.一种基于小波变换的机动目标跟踪算法的研究[J].系统工程与电子技术,2002,(6):60-63.
[99]Karaulova I, Hall P, Marshall A. A hierarchical model of dynamics for tracking peoplewith a single video camera [C]. ICCV.2000,352-361.
[100] Wachter S, Nagel H. Tracking of persons in monocular image sequences [C]. InNonrigid and Articulated Motion Workshop.1997,2-9.
[101] Ju S, Black M, Yacoob Y. Cardboard people:A parameterized model of articulatedimage motion [J]. Automatic Face and Gesture Recognition,2002,38-44.
[102] R S Busy, K D Renne. Digital Synthesis of Nonlinear Filter [J].Automatica,1971,7(3):287-289.
[103] T Yoshimura, T Sueda. A Technique for Compensating the Filter Performance by aFictitious Noise [J]. Trans.ASME,1978,100(2):154-156.
[104] A P Sage, G W Husa. Adaptive Filtering with Unknown Prior Statistics [C]. JointAutomatic Control Conference. Boulder,1969:760-769.
[105] B Friedland. Treatment of Bias in Recursive Filtering [J]. IEEE Trans.AC,1969,14(4):359-362.
[106] Ristic B, Arulampalam S, Gordon N. Beyond the Kalman Filter: Particle Filters forTracking Applications[M]. London:Artech House,2004.
[107] SUN Kai, LIU Shi-rong. Combined algorithm with modified Camshift and KalmanFilter for multi-object tracking [J]. Information and Control,2009,38(1):11-16.
[108] Adini Y, Moses Y, Ullman S. FaeeReeognition:The Problem of Compensating forChanges in illumination Direction [J].IEEE Transactions on Pattern Analysis andMachine Intelligence,1997,19(7):721-732.
[109] Grother P, Micheals R, PhilliPs P. Face recognition vendor test2002Performancemetrics [R].2003,1057-1059.
[110] W W Bledsoe. The model method in facial recognition [R], Technical report PR115,Panoramic Research Inc,1966.
[111] Junfeng Yao, Yu Zhan, Chih-Cheng Hung, Xiaobiao Xie, Guoli Ji. Splicing Side andFront Face Images Seamlessly by Applying Iterative Fusion Algorithm with n*nWave-low-pass Filters [J]. Information,2011,14:609-620.
[112] Miao Song, Keizo Shinomori, Shiyong Zhang. How do Facial Parts Contribute toExpression Perception? An Answer from the High-level Face Adaptation [J].Information,2010,1947-1956.
[113] Zhao-yi Peng, Yu Zhou, Yan-hui Zhu, Zhi-qiang Wen. Application of an ImprovedMean Shift Algorithm in Real-time Facial Expression Recognition [J]. Journal ofSoftware,2011,6:100-107.
[114] huanxu Wang. Face Segmentation Based on Skin Color in Complicated Backgroundand Its Sex Recognition [J]. Journal of Software,2011,1209-1216.
[115] S D Chen, A R Ramli. Minimum Mean Brightness Error Bi-Histogram Equalization inContrast Enhancement [J]. IEEE Transactions on Consumer Electronics,2003,49:1310-1319.
[116] W C Kao, M C Hsu, Y Y Yang. Local contrast enhancement and adaptive featureextraction for illumination-invariant face recognition [J]. Pattern Recognition,2010,43:1736-1747.
[117] Chengjun Liu, Harry Wechsler. Evolutionary Pursuit and Its Application to FaceRecognition [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(6):570-582.
[118] R Brunelli, T Poggio. HyperBF Networks for Gender Classification [C]. In Proceedingof DARPA Image Understanding Workshop.1992,311-314.
[119] Stev Lawrence, C Lee Giles, Ah Chung Tsoi. Face Recognition: A Convolution NeuralNetwork Approach [J]. IEEE Transactions on Neural Networks,1997,8(1):98-113.
[120]刘小军.人脸识别技术研究[D].中国科学院.2001.
[121]李士进,杨静宇,陆建峰.基于奇异值特征和隐马尔可夫模型的人脸检测[J].中国图象图形学报,2001,6(7):681-687.
[122] Volker Blanz, Thomas Vetter. Face Recognition Based on Fitting a3D MorphableModel [J]. IEEE transactions on Pattern Analysis and Machine Intelligence,2003,25(9):1063-1075.
[123] G Donato, M S Bartlett, J C Hager. Classifying Facial Actions [J]. IEEE Trans. PatternAnalysis and Machine Intelligence,1999,21(10):974-989.
[124] Jian Yang, Jing-Yu Yang, From image vector to matrix:a straightforward imageprojection technique-IMPCA vs. PCA [J]. Pattern Recognition,2002,35:1997-1999.
[125] Jafri R，Arabnia H R．A Survey of Face Recognition Techniques [J]．Journal ofInformation Processing Systems,2009,5(2):41-68.
[126] ZHANG Chao-zhu, ZUO Guo-hui. Kernel Methods in Face Recognition Based on2DLDA [J]. Journal of Jilin University,2011,41(4):1167-1170.
[127] LI Ying, GONG Hong-li, LIANG Jia-xi. SAR Image Target Feature Extraction Basedon KSVD and PCA [J]. Journal of Jilin University,2010，40(5):1336-1339.
[128] YANG Qiong, DING Xiao-qing. Symmetrical PCA and Its Application to FaceRecognition [J]. Chinese Journal of Computers,2003,26(9):1146-1151.
[129]高璐.基于模糊聚类分析和多特征融合的人脸识别方法[J].吉林大学学报,2012,50(2),293-298.
[130] Comaniciu D, Ramesh V, Meer P. Real-time tracking of non-rigid objects using meanshift [J]. Computer Vision and Pattern Recognition,2000:142-149.
[131] PENG Ningsong, YANG Jie, LIU Zhi, ZHANG Fengchao. Automatic selection ofkernel-bandwidth for mean-shift object tracking [J]. Journal of Software,2005,16(9):1542-1550.
[132]夏波涌,童悍操.基于形态滤波和均值漂移红外小目标检测方法[J].计算机仿真.2009,2,15-19.
[133] G. Hua, A Akbarzadeh. A robust elastic and partial matching metric for face recognition
[C]. ICCV.2009.
[134] E M Oehoa. Clustered microcalcification detection using optimized difference ofGaussians [J]. Air Force institute of Technology,1996.
[135] C.Stauffer,W E L Grimson. adaptive background mixture models for real-timetracking[C].CVPR.1999,246-252.
[136] Zhichao Lian, Meng Joo Er, Juekun Li. A Novel Face Recognition Approach underIllumination Variations Based on Local Binary Pattern [C]. In Proceeding of the14thinternational conference on Computer analysis of images and patterns.2011,6855,89-96.
[137] E Izquierdo, M Ghanbari. Nonlinear gaussian filtering approach for object segmentation[J]. Vision, Image and Signal Processing,1999,146:137-143.
[138] Weina Fu, Zhiwen Xu, Shuai Liu, Xin Wang, Hongchang Ke. The capture of movingobject in video image [J]. Journal of Multimedia,2011,6:518-525.
[139]周杰,卢春雨,张长水,李衍达.人脸自动识别方法综述[J].电子学报,2000,28(4):102-106。
[140] Hanyang Jiang. Research on the Document Image Segmentation Based on the LDAModel [J]. Advances in Information Sciences and Service Sciences,2012,4(3):12-18.
[141] Seyed Zeinolabedin Moussavi, Saeedreza Ehteram, Ali Sadr. The New FaceRecognition Technique With the use of PCA and LDA [J]. Journal of ConvergenceInformation Technology,2008,3(2):34-38.
[142] Zhou Jie, Lu Chunyu, Zhang Changshui, Li Yanda. A Survey of Automatic Human FaceRecognition [J]. ACTA ELECTRONICA SINICA,2000,28(4):102-106.
[143] Huilin Xiong, M N S Swamy, M Omair Ahmad. Optimizing the kernel in the empiricalfeature space [J]. IEEE Transactions on Neural Networks, Institute of Electrical andElectronics Engineers,2005,16(2):460-474.
[144] Jerome Friedman, Trevor Hastie, Robert Tibshirani. Additive logistic regression: Astatistical view of boosting [J]. Annals of Statistics,2000,28(2):337-407.
[145] Juwei Lu, K N Plataniotis, A N Venetsanopoulos, Stan Z Li. Ensemble-baseddiscriminant learning with boosting for face recognition [J]. IEEE Transactions onNeural Networks,2006,17(1):166-178.
[146] Dao-Qing Dai, PC Yuen. Face recognition by regularized discriminant analysis [J].IEEE Transactions on Systems, Man, and Cybernetics,2007,37(4):1080-1085.
[147] Jieping Ye, Qi Li. A two-stage linear discriminant analysis via QR-decomposition [J].Pattern Analysis and Machine Intelligence,2005,27(6):929-941.
[148] Jun Liu, Songcan Chen. Discriminant common vectors versus neighbourhoodcomponents analysis and laplacian faces: A comparative study in small sample sizeproblem [J]. Image and Vision Computing,2006,24(3):249-262.
[149] XiaoGang Wang, Xiaoou Tang. Random sampling for subspace face recognition [J].International Journal of Computer Vision,2006,70(1):91-104.
[150] Dijun Luo, Chris Ding, Heng Huang. Symmetric two dimensional linear discriminantanalysis (2DLDA)[C]. In Proceedings of IEEE Conference on Computer Vision andPattern Recognition.2009,2820-2827.
[151] Xiao-ming Wang, Chang Huang, Jin-gao Liu. Gabor-2DLDA: face recognition usingGabor features and2D linear discriminant analysis [C]. In Proceedings of the SecondInternational Conference on Intelligent Computation Technology andAutomation,.2009,608-610.
[152] Gui-Fu Lu, Jian Zou, Yong Wang. Incremental complete LDA for face recognition [J].Pattern Recognition,2012,45(7):2510-2521.
[153] SenJian An, Wanquan Liu, Svetha Venkatesh, Hong Yan. Unified formulation oflinear discriminant analysis methods and optimal parameter selection [J].2011,44(2):307-319.
[154] AM Martinez, R Benavente. The AR Face Database [R]. Computer Vision Center(CVC),Spain,1998.
[155]王莹.基于图像的人脸识别技术研究[D].吉林大学.2012.
[156]李文辉,姜园媛,王莹,傅博.一种基于重采样双向2DLDA融合的人脸识别算法[J],电子学报.2011.
[2] Marr D.姚国正等译.视觉计算理论[M].北京：科学出版社,1988.
[3]崔之枯，江春，陈丽鑫.数字视频处理[M].北京:电子工业出版社,1998.
[4]刘小军.人脸识别技术研究[D].中国科学院.2001.
[5] Prange J. Detecting, recognizing and understanding video events in surveillancevideo[J].Advanced Video and Signal Based Surveillance,2003,4.
[6]王亮,胡卫明,谭铁牛．人运动的视觉分析综述[J]．计算机学报，2002,25(3):1-16.
[7]边肇祺,张学工.《模式识别》第二版[M].北京：清华大学出版社,2001：223-227.
[8] D.M.Gavrila. The Visual Analysis of Human Movement: A Survey. Computer Vision andImage Understanding.1999,73(1):82-98.
[9] Ismail Haritaoglu,David Harwood,Larry S.Davis.W4:Real-Time Surveillance of Peopleand Their Activities[J].IEEE Transactions on Pattern Analysis and MachineIntelligence.2000,22(8):809-830.
[10]Matsuyama T, Ukita N. Real-time multitarget tracking by a cooperative distributedvision system[J].Proceedings of the IEEE,2002,90:1136-1150.
[11]Shu C, Hampapur A, Lu M. IBM smart surveillance system (S3):an open and extensiblearchitecture for smart video surveillance[C]. In IEEE International Conference onAdvanced Video and Signal based Surveillance,2005,318-323.
[12]Harmon L D. The recognition of faces[J].Sci.1973,229:71-82.
[13]T Kanade. Picture processing system by computer and recognition of human faces[D].Kyoto University,1973.
[14]Baron RJ. Mechanisms of human facial recognition [J]. Man-machine Studies.1981,15:137-178.
[15]Alex Pentland, Baback Moghaddam, Thad starner. View-Based and ModularEigen-spaces for Face Recognition[C]. MIT Media Lab perceptual Computing SectionTech IEEE Conference on Computer Vision&Pattern Recognition,1994.
[16]P N Bellhumer, J Hespanha, D Kriegman. Eigenfaces vs.fisherfaces: Recognition usingclass specific linear projection [J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1997,17(7):711-720.
[17]W.Zhao,R.Chellappa. SFS Based View Synthesis for Robust Face Recognition[C], InProceeding of the4th International Conference on Face and Gesture Recognition,2000,285-292.
[18]C Liu, H Wechsler. Gabor Feature Based Classification Using the Enhanced Fisher LinearDiscriminant Model for Face Recognition [J].IEEE Trans.Image Processing.2002,11(4):467-476.
[19]H Yu, J Yang. A Direct LDA Algorithm for High-dimensional Data–with Application toFace Recognition[J]. Pattern Recognition.2001,34:2067-2070.
[20]刘青山.人脸跟踪与识别的研究[D].中科院自动化所.2003.
[21]Ming Li，Baozong Yuan．2D-LDA:A Statistical Linear Discriminant Analysis for ImageMatrix[J]．Pattern Recognition Letters．2005,26(5):527-532.
[22]B.Moghaddam, A.Pentland. Probabilistic Visual Learning for Object representation [J].IEEE Trans.On PAMI.1997,20(7):696-710.
[23]Baback Moghaddam, Tony Jebara, Alex Pentland. Bayesian Face Recognition [J]. PatternRecognition.2000,33:1771-1782.
[24]J Buhmann, M Lades, von der Malsburg. Size and distortion invariant object recognitionby hierarchical graph matching[C]. In Proceedings of IEEE Intl. Joint Conference onNeural Networks. Santiago,1990,411-416.
[25]A Tefas, C Kotropoulos, I Pitas. Using Support Vector Machines to Enhance thePerformance of Elastic Graph Matching for Frontal Face Authentication [J]. IEEETransactions On Pattern Analysis And Machine Intelligence.2001,23(7):735-746.
[26]丁嵘,苏光大,林行刚.使用关键点信息改进弹性匹配人脸识别算法[J].电子学报.2002,30(9):1292-1294.
[27]Hochul Shin, Seong Dae Kim, Hae Chul Choi. Generalized elastic graph matching forface recognition [J]. Pattern Recognition Letters.2007,28(9):1077-1082.
[28]P.Penev, J.Atick. Local Feature Analysis: A General Statistical Theory for ObjectRepresentation [J]. Computation in Neural Systems.1996,7:477-500.
[29]A Lanitis, C J Taylor, T F Cootes. An automatic face identification system using flexibleappearance models [C]. In British Machine Vision Conference.1994,65-74.
[30]A Lanitis, C Taylor, T Cootes. A Unified Approach to Coding and Interpreting FaceImages [C]. In Proceedings of Int'l Conference Computer Vision.1995,368-373.
[31]T F Cootes, G J Edwards, C J Taylor. Active Appearance Models [C], In Proceedings ofEuropean Conference Computer Vision.1998,484-498.
[32]G Edwards, T Cootes, C Taylor. Advances in Active Appearance Models, In Proceedingsof Int'l Conference Computer Vision.1999,137-142.
[33]M B Stegmann. Active Appearance Models, Theory, Extension and Cases, Master thesis[D]. Technical University of Denmark.2000.
[34]S C Yan, C Liu, S Z Li, H J Zhang, H Shum,Q S Cheng. Face Alignment UsingTexture-Constrained Active Shape Models [J]. Image and Vision Computing.2003,21(1):69-75.
[35]Tong Wang, Haizhou Ai, Gaofeng Huang. A Two-Stage Approach to Automatic FaceAlignment [C]. In Proceedings of SPIE-The International Society for Optical Engineering.2003,558-563.
[36]Yi Zhou, Lie Gu, Hongjiang Zhang. Bayesian Tangent Shape Model: Estimating Shapeand Pose Parameters via Bayesian Inference [C]. In Proceeding of The IEEE Conferenceon Computer Vision and pattern Recognition. USA: Wisconsin,2003,16-22.
[37]Shiguang Shan, Wen Gao, Wei Wang, Debin Zhao, Baocai Yin. Enhanced Active ShapeModels with Global Texture Constraints for Face Image Analysis[C]. FourteenthInternational Symposium On Methodologies For Intelligent Systems. Japan: MaebashiCity,2003,593-597.
[38]Fatih Kahraman, Mikkel B Stegmann. Towards Illumination-invariant Localization ofFaces using Active Appearance Models [C]. The7th Nordic Signal ProcessingSymposium. Iceland: Reykjavik,2006.
[39]P J Philips, P J Grother, R J Micheals, D M Blackburn, E Tabassi, J M Bone. FaceRecognition Vendor Test2002: Evaluation report. Technical Report. National Institute ofStandards and Technology.2003.
[40]G Guo, S Z Li, K Chan. Face Recognition by Support Vector Machines [C]. InProceedings of the4th International Conference on Auto. Face and Gesture Recognition.France: Grenoble,2000,196-201.
[41]P J Phillips. Support vector machines applied to face recognition [C]. In Advances inNeural Information Processing Systems.1998,803.
[42]Volker Blanz, Thomas Vetter. A Morphable Model For the Synthesis of3D Faces [J].SIG’ GRAPH’,1999.
[43]Sami Romdhani, Volker Blanz, Thomas Vetter. Face Identification by Fitting a3DMorphable Model Using Linear Shape and Texture Error Functions [C]. In Proceedingsof the7th European Conference on Computer Vision,2002,3-19.
[44]Y Freund, R E Schapire. A decision-theoretic generalization of on-line learning and anapplication to boosting [J]. Journal of Computer and System Sciences,,1997,55(1):119-139.
[45]R Meir, G Raetsch. An introduction to Boosting and Leveraging [J]. Advanced Lectureson Machine Learning,2003:118-183.
[46]J Lu, KN Plataniotics, A N Venetsanopoulos. Ensemble-based discriminant learning withboosting for face recognition [J]. IEEE transactions on neural networks,2006,17(1):166-178.
[47]张文超,山世光,张洪明,陈杰,陈熙霖,高文.基于局部Gabor变化直方图序列的人脸描述与识别[J].软件学报,2006,17(12):2508-2517.
[48]Xiangsheng Huang, Stan Z Li, Wang Yangsheng. Jensen-shannon boosting for objectrecognition [C]. International conference on computer vision and patternrecognition,2005.
[49]A S Georghiades, D J Kriegman, P N Belhumeur. Illumination Cones For RecognitionUnder Variable Lighting: Faces. In Proceedings of IEEE CVPR,1998,52-58.
[50]A Shashua, T Riklin Raviv. The Quotient Image: Class-Based Re-Rendering AndRecognition With Varying Illuminations [J]. IEEE Trans. on PAMI,2001,23(2):129-139.
[51]R Basri, D Jacobs. Lambertian Reflectance and Linear Subspaces [C],ICCV2001.Beckman Institute,2001,383-390.
[52]张志伟.可变光照条件下的人脸识别技术研究[D].河北工业大学.2007
[53]A R Martinez, R Benavente. The AR face database. Computer VisionCenter(CVC)Technical Report,1998.
[54]王先基.基于统计学习的自动人脸识别算法研究[D].中国科学技术大学.2007.
[55]卢官明,毕厚杰.基于数学形态学的图像序列分割[J].南京邮电学院学报,1997,17(2):54-57.
[56]韩军,熊琼,孙文彦等.自动分割及跟踪视频运动对象的一种实现方法[J].中国图像图形学报,2001,(8):732-738.
[57]Laurenz Wiskott. Segmentation from Motion:Combining Gabor-and Mallat-Wavelets toOvercome the Aperture and Correspondence Problems [J]. PatternRecognition,1999:1751-1766.
[58]黄友珍,黄艺,余兆明.基于修正分水岭算法和时域跟踪的视频自动分割[J].数字视频,2000:5-8.
[59]Gerald Kuhne, Stephan Richter, Markus Beier. Motion-Based Segmentation andContour-Based Classification of Video Objects [J]. ACM Multimedia,2001,6(9):254-273.
[60]黄波,杨勇,王桥等.一种用于视频分割德快速运动估计方法[J].电路与系统学报,2001,6(1):69-71.
[61]Y.Kim, H.Joon. Model-based Discontinuity Evaluation in the DCT Domain [J]. SignalProcessing,2001,81(4):871-882.
[62]詹劲峰,戚飞虎,王海龙.基于时空马尔可夫随机场德运动目标分割技术[J].通信学报,2000,21(11):63-83.
[63]Kuo Chung-Ming, Hsieh Chaur-Heh, Huang Yong-Ren. A New Mesh Based onTemporal-Spatial Segmentation for Image Sequence [C]. Computer Software andApplications Conference,2000:196-222.
[64]吴枫,高鹏,高文.基于网格模型的运动估计技术[J].电子学报,2000,28(5):47-51.
[65]夏利民,古士文.基于形变模型的图像分割和非刚性目标的跟踪[J].模式识别与人工智能,2000,13(1):64-68.
[66]Harper H.s, Dinggang shen. An Affine-Invariant Active Contour Model for Medel-BasedSegmentation [J]. Image and Vision Computing,1998,(16):135-146.
[67]Fan Jianping,Yu,Jun Gen Fujita. Spatio-Temporal Segmentation for Compact VideoRepresentation [J]. Signal Processing:Image Communication,2001,(16):553-566.
[68]施可为,傅锡天,蔡安妮等.含人脸的前景再活动图像序列中的分割[J].北京邮电大学学报,2000,23(1):66-70.
[69]张颖,张兆阳.基于二次空间变换的快速时空分割[J].中国图像图形学报,2000,5(9):744-749.
[70]Laurenz Wiskott. Segmentation from Motion:Combining Gabor-and Mallat-Wavelets toOvercome the Aperture and Correspondence Problems [J]. PatternRecognition,1999:1751-1766.
[71]Daniel Gatica-Perez, Sun,Ming-Ting Chuang Gu. Multlview Extensive PartitionOperators for Semantic Video Object Extraction [J]. IEEE Transactions on Circuits andSystems for Video Technology,2002,11(7):788-800.
[72]俞刚毅,黄艺.用于视频对象平面生成的运动对象自动分割[J].电子与信息学报,2001,23(7):657-662.
[73]N Friedman, S Russell. Image Segmentation In Video Sequences:A ProbabilisticApproach Uncertainty [J]. Artificial Intelligence.1997.
[74]A Elgammal, D Harwood, L Davis. Non-Parametric Model for Background Subtraction
[C]. In Proceedings IEEE Frame Rate Workshop.1999.
[75]Ismail Haritaoglu, David Harwood, Larry S.Davis. W4:Real-Time Surveillance of Peopleand Their Activities [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):809-830.
[76]ZENG Chun, LI Xiao-hua, ZHOU Ji-liu. Pedestrian detection based on HOG of ROI [J].Computer Engineering,2009,35(24):182-184.
[77]Lin Rui，Du Zhijiang，Sun Lining. Moving Object Tracking based on Mobile RobotVision [C], In Proceedings of International Conference on Mechatronics and Automation.2009.
[78]栾大勇.实时视频监控系统中的运动人体检测算法研究[D].吉林大学.2009.
[79]L.Wixson. Detecting Salient Motion by Accumulating Directionally-Consistent Flow[J].IEEE transaction on Pattern Analysis and Machine Intelligence,2000,22(8).:774-781.
[80]A.Shio, J Sklansky. Segmentation of people in motion [C]. In Proceeding IEEEWorkshop on Visual Motion,1991,325-332.
[81]Nuno Vasconcelos, Andrew Lippman. Empirical Bayesian Motion Segmentation [J].IEEE transactions on Pattern Analysis and Machine Intelligence,2001,23(2):217-221.
[82]WANG Xiaoyan，ZHANG Yanzhu，DONG Huiying，LI Yuan，LI Xiaojuan. AlgorithmResearch on Three-Frame Difference for Detection of Moving Target [J]. Journal ofShenyang Ligong University,2011,30(6):82-85.
[83]Navneet Dalal, Bill Triggs. Histograms of Oriented Gradients for Human Detection[C].In Proceedings of CVPR2005.
[84]Q Zhu, S Avidan, M C Yeh, K T Cheng. Fast Human Detection Using a Cascadeof Histogram of Oriented Gradients[C].In Proceedings of IEEE International Conference onComputer Vision and Pattern Recognition.2006.
[85]Bo Wu, Ram Nevatia. Detection of Multiple, Partially Occluded Humans in a SingleImage by Bayesian Combination of Edgelet Part Detectors[C]. In Proceedings ofICCV.2005.
[86]P Sabzmeydani, G Mori. Detecting pedestrians by learning shapelet features[C]. InProceedings of CVPR.2007.
[87]倪洪印.基于视频序列的人体检测和跟踪技术的研究[D].吉林大学.2010.
[88]刘海松.基于广义互补编码的目标识别处理器[J].中国激光，1999,26(3):1031-1035.
[89]Munchurl Kim, Jeon J G, Kwak J.S. Moving Object Segmentation in Video Sequences byUser Interaction and Automatic Object Tracking [J]. Image and Vision Computing,2001,19(5):245-258.
[90]Candemir Toklu, A Murat Tekalp, Tanju Erdem. Semi-Automatic Video Segmentation inthe Presence of Occlusion [J]. IEEE Transactions on Circuits and Systems for VideoTechnology,2002,10(4):624-629.
[91]Christopher Rasmussen, Gregory D.Hager. Probabilistic Data Association Method forTracking Complex Visual Objects [J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,2000,23(6):560-576.
[92]Holger Eggers, Fabrice Moscheni, Poberto Castagno. Robust Object Tracking Based onSpatial Characterization of Objects by Additive Invariant [J]. IEEE Transactions onCircuits and Systems for Video Technology,1999,9(9):844-848.
[93]Francois Bremond, Monique Thonnat. Tracking Multiple Non-Rigid Objects in VideoSequences [J]. IEEE Transactions on Circuits and Systems for VideoTechnology,1998,8(5):585-591.
[94]Zhang Jun, Gao Jianbo, Liu Wei-song. Image Sequence Segmentation Using3-DStructure Tensor and Curve Evolution [J]. IEEE Transactions on Circuits and Systems forVideo Technology,2001,11(5):629-641.
[95]Gerald Kuhne, Stephan Richter, Markus Beier. Motion-Based Segmentation andContour-Based Classification of Video Objects [J]. ACM Multimedia,2001,6(9):254-273.
[96]施建良,余松煜,傅立言.基于最短描述长度的序列图像运动分割[J].计算机学报,2001,22(8):809-815.
[97]Donna J.Williams, Mubarak Shah. A Fast Algorithm for Active Contours and CurvatureEstimation [J]. Image Understanding,1992,1(55):14-26.
[98]郭春,罗鹏飞.一种基于小波变换的机动目标跟踪算法的研究[J].系统工程与电子技术,2002,(6):60-63.
[99]Karaulova I, Hall P, Marshall A. A hierarchical model of dynamics for tracking peoplewith a single video camera [C]. ICCV.2000,352-361.
[100] Wachter S, Nagel H. Tracking of persons in monocular image sequences [C]. InNonrigid and Articulated Motion Workshop.1997,2-9.
[101] Ju S, Black M, Yacoob Y. Cardboard people:A parameterized model of articulatedimage motion [J]. Automatic Face and Gesture Recognition,2002,38-44.
[102] R S Busy, K D Renne. Digital Synthesis of Nonlinear Filter [J].Automatica,1971,7(3):287-289.
[103] T Yoshimura, T Sueda. A Technique for Compensating the Filter Performance by aFictitious Noise [J]. Trans.ASME,1978,100(2):154-156.
[104] A P Sage, G W Husa. Adaptive Filtering with Unknown Prior Statistics [C]. JointAutomatic Control Conference. Boulder,1969:760-769.
[105] B Friedland. Treatment of Bias in Recursive Filtering [J]. IEEE Trans.AC,1969,14(4):359-362.
[106] Ristic B, Arulampalam S, Gordon N. Beyond the Kalman Filter: Particle Filters forTracking Applications[M]. London:Artech House,2004.
[107] SUN Kai, LIU Shi-rong. Combined algorithm with modified Camshift and KalmanFilter for multi-object tracking [J]. Information and Control,2009,38(1):11-16.
[108] Adini Y, Moses Y, Ullman S. FaeeReeognition:The Problem of Compensating forChanges in illumination Direction [J].IEEE Transactions on Pattern Analysis andMachine Intelligence,1997,19(7):721-732.
[109] Grother P, Micheals R, PhilliPs P. Face recognition vendor test2002Performancemetrics [R].2003,1057-1059.
[110] W W Bledsoe. The model method in facial recognition [R], Technical report PR115,Panoramic Research Inc,1966.
[111] Junfeng Yao, Yu Zhan, Chih-Cheng Hung, Xiaobiao Xie, Guoli Ji. Splicing Side andFront Face Images Seamlessly by Applying Iterative Fusion Algorithm with n*nWave-low-pass Filters [J]. Information,2011,14:609-620.
[112] Miao Song, Keizo Shinomori, Shiyong Zhang. How do Facial Parts Contribute toExpression Perception? An Answer from the High-level Face Adaptation [J].Information,2010,1947-1956.
[113] Zhao-yi Peng, Yu Zhou, Yan-hui Zhu, Zhi-qiang Wen. Application of an ImprovedMean Shift Algorithm in Real-time Facial Expression Recognition [J]. Journal ofSoftware,2011,6:100-107.
[114] huanxu Wang. Face Segmentation Based on Skin Color in Complicated Backgroundand Its Sex Recognition [J]. Journal of Software,2011,1209-1216.
[115] S D Chen, A R Ramli. Minimum Mean Brightness Error Bi-Histogram Equalization inContrast Enhancement [J]. IEEE Transactions on Consumer Electronics,2003,49:1310-1319.
[116] W C Kao, M C Hsu, Y Y Yang. Local contrast enhancement and adaptive featureextraction for illumination-invariant face recognition [J]. Pattern Recognition,2010,43:1736-1747.
[117] Chengjun Liu, Harry Wechsler. Evolutionary Pursuit and Its Application to FaceRecognition [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(6):570-582.
[118] R Brunelli, T Poggio. HyperBF Networks for Gender Classification [C]. In Proceedingof DARPA Image Understanding Workshop.1992,311-314.
[119] Stev Lawrence, C Lee Giles, Ah Chung Tsoi. Face Recognition: A Convolution NeuralNetwork Approach [J]. IEEE Transactions on Neural Networks,1997,8(1):98-113.
[120]刘小军.人脸识别技术研究[D].中国科学院.2001.
[121]李士进,杨静宇,陆建峰.基于奇异值特征和隐马尔可夫模型的人脸检测[J].中国图象图形学报,2001,6(7):681-687.
[122] Volker Blanz, Thomas Vetter. Face Recognition Based on Fitting a3D MorphableModel [J]. IEEE transactions on Pattern Analysis and Machine Intelligence,2003,25(9):1063-1075.
[123] G Donato, M S Bartlett, J C Hager. Classifying Facial Actions [J]. IEEE Trans. PatternAnalysis and Machine Intelligence,1999,21(10):974-989.
[124] Jian Yang, Jing-Yu Yang, From image vector to matrix:a straightforward imageprojection technique-IMPCA vs. PCA [J]. Pattern Recognition,2002,35:1997-1999.
[125] Jafri R，Arabnia H R．A Survey of Face Recognition Techniques [J]．Journal ofInformation Processing Systems,2009,5(2):41-68.
[126] ZHANG Chao-zhu, ZUO Guo-hui. Kernel Methods in Face Recognition Based on2DLDA [J]. Journal of Jilin University,2011,41(4):1167-1170.
[127] LI Ying, GONG Hong-li, LIANG Jia-xi. SAR Image Target Feature Extraction Basedon KSVD and PCA [J]. Journal of Jilin University,2010，40(5):1336-1339.
[128] YANG Qiong, DING Xiao-qing. Symmetrical PCA and Its Application to FaceRecognition [J]. Chinese Journal of Computers,2003,26(9):1146-1151.
[129]高璐.基于模糊聚类分析和多特征融合的人脸识别方法[J].吉林大学学报,2012,50(2),293-298.
[130] Comaniciu D, Ramesh V, Meer P. Real-time tracking of non-rigid objects using meanshift [J]. Computer Vision and Pattern Recognition,2000:142-149.
[131] PENG Ningsong, YANG Jie, LIU Zhi, ZHANG Fengchao. Automatic selection ofkernel-bandwidth for mean-shift object tracking [J]. Journal of Software,2005,16(9):1542-1550.
[132]夏波涌,童悍操.基于形态滤波和均值漂移红外小目标检测方法[J].计算机仿真.2009,2,15-19.
[133] G. Hua, A Akbarzadeh. A robust elastic and partial matching metric for face recognition
[C]. ICCV.2009.
[134] E M Oehoa. Clustered microcalcification detection using optimized difference ofGaussians [J]. Air Force institute of Technology,1996.
[135] C.Stauffer,W E L Grimson. adaptive background mixture models for real-timetracking[C].CVPR.1999,246-252.
[136] Zhichao Lian, Meng Joo Er, Juekun Li. A Novel Face Recognition Approach underIllumination Variations Based on Local Binary Pattern [C]. In Proceeding of the14thinternational conference on Computer analysis of images and patterns.2011,6855,89-96.
[137] E Izquierdo, M Ghanbari. Nonlinear gaussian filtering approach for object segmentation[J]. Vision, Image and Signal Processing,1999,146:137-143.
[138] Weina Fu, Zhiwen Xu, Shuai Liu, Xin Wang, Hongchang Ke. The capture of movingobject in video image [J]. Journal of Multimedia,2011,6:518-525.
[139]周杰,卢春雨,张长水,李衍达.人脸自动识别方法综述[J].电子学报,2000,28(4):102-106。
[140] Hanyang Jiang. Research on the Document Image Segmentation Based on the LDAModel [J]. Advances in Information Sciences and Service Sciences,2012,4(3):12-18.
[141] Seyed Zeinolabedin Moussavi, Saeedreza Ehteram, Ali Sadr. The New FaceRecognition Technique With the use of PCA and LDA [J]. Journal of ConvergenceInformation Technology,2008,3(2):34-38.
[142] Zhou Jie, Lu Chunyu, Zhang Changshui, Li Yanda. A Survey of Automatic Human FaceRecognition [J]. ACTA ELECTRONICA SINICA,2000,28(4):102-106.
[143] Huilin Xiong, M N S Swamy, M Omair Ahmad. Optimizing the kernel in the empiricalfeature space [J]. IEEE Transactions on Neural Networks, Institute of Electrical andElectronics Engineers,2005,16(2):460-474.
[144] Jerome Friedman, Trevor Hastie, Robert Tibshirani. Additive logistic regression: Astatistical view of boosting [J]. Annals of Statistics,2000,28(2):337-407.
[145] Juwei Lu, K N Plataniotis, A N Venetsanopoulos, Stan Z Li. Ensemble-baseddiscriminant learning with boosting for face recognition [J]. IEEE Transactions onNeural Networks,2006,17(1):166-178.
[146] Dao-Qing Dai, PC Yuen. Face recognition by regularized discriminant analysis [J].IEEE Transactions on Systems, Man, and Cybernetics,2007,37(4):1080-1085.
[147] Jieping Ye, Qi Li. A two-stage linear discriminant analysis via QR-decomposition [J].Pattern Analysis and Machine Intelligence,2005,27(6):929-941.
[148] Jun Liu, Songcan Chen. Discriminant common vectors versus neighbourhoodcomponents analysis and laplacian faces: A comparative study in small sample sizeproblem [J]. Image and Vision Computing,2006,24(3):249-262.
[149] XiaoGang Wang, Xiaoou Tang. Random sampling for subspace face recognition [J].International Journal of Computer Vision,2006,70(1):91-104.
[150] Dijun Luo, Chris Ding, Heng Huang. Symmetric two dimensional linear discriminantanalysis (2DLDA)[C]. In Proceedings of IEEE Conference on Computer Vision andPattern Recognition.2009,2820-2827.
[151] Xiao-ming Wang, Chang Huang, Jin-gao Liu. Gabor-2DLDA: face recognition usingGabor features and2D linear discriminant analysis [C]. In Proceedings of the SecondInternational Conference on Intelligent Computation Technology andAutomation,.2009,608-610.
[152] Gui-Fu Lu, Jian Zou, Yong Wang. Incremental complete LDA for face recognition [J].Pattern Recognition,2012,45(7):2510-2521.
[153] SenJian An, Wanquan Liu, Svetha Venkatesh, Hong Yan. Unified formulation oflinear discriminant analysis methods and optimal parameter selection [J].2011,44(2):307-319.
[154] AM Martinez, R Benavente. The AR Face Database [R]. Computer Vision Center(CVC),Spain,1998.
[155]王莹.基于图像的人脸识别技术研究[D].吉林大学.2012.
[156]李文辉,姜园媛,王莹,傅博.一种基于重采样双向2DLDA融合的人脸识别算法[J],电子学报.2011.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |