基于均值偏移算法的运动目标跟踪算法的研究
|
摘要
视频目标跟踪是计算机视觉的核心技术之一，它融合了图像处理、自动控制和人工智能等众多领域，其在视频监控、医疗诊断、生物学研究、人机交互、自动驾驶及机器人等领域具有重要的科学理论意义和工程应用价值。本文以Mean Shift (MS)作为目标跟踪的核心算法，以提高目标跟踪的准确性、鲁棒性和适应性为目标，旨在解决视频目标跟踪中部分遮挡、形变、光照变化、尺度变化和跟踪矩形可调整姿态等问题。
传统颜色直方图的MS算法只考虑了目标颜色的统计信息，不包含目标的空间信息，当目标颜色与背景颜色相近时，容易导致不准确跟踪或跟踪丢失。针对该问题，提出了一种自适应空间颜色直方图的MS跟踪算法。该算法根据目标对象的最新外接矩形尺寸，确定对象分块方法，根据各块的Bhattacharyya系数值，确定各块的权重系数。其中，自适应分块的颜色直方图包含了自适应分块方法和目标的空间信息；加权Bhattacharyya系数考虑到不同块对整体相似度的不同影响。研究表明，提出的算法可以动态地调整跟踪矩形水平和垂直方向的尺度，自适应地确定对象的分块方法，在部分遮挡和形变等情况下，比传统的MS算法和固定分块的MS算法具有更好的跟踪性能。
考虑到传统MS算法易受光照变化、部分遮挡、图像模糊等因素的影响，引入了图像局部特征点。采用实验对比的方法从旋转不变性、尺度不变性、光照不变性、抗模糊性、跟踪精度及提取时间等六个方面，对目前最流行的六个特征点算法及描述子(BRIEF、LAZY、ORB、RIFF、SIFT和SURF)等进行比较，最终确定了SIFT作为本文采用的特征点及描述子。
针对传统MS算法易受光照变化、部分遮挡和图像模糊等因素影响的问题，提出了一种融合改进MS和SIFT的跟踪算法，该方法由改进的MS跟踪算法(初定位)和SIFT特征提取、匹配和跟踪(SIFT跟踪)组成。前者即第三章算法，后者在前者跟踪结果的基础上，采用SIFT跟踪方法得到SIFT跟踪结果。最后，通过线性加权的方法融合改进MS和SIFT的跟踪结果，获得最终的跟踪结果。研究表明，提出的算法进一步地解决了部分遮挡、光照变化和图像模糊情况下的目标跟踪问题。
为了解决目标对象跟踪矩形可自适应调整姿态的问题，提出了一种融合MS和SIFT的仿射变换目标跟踪算法。将仿射变换引入到目标的候选模型中，将复杂运动跟踪问题转化为代价函数的优化问题。通过对代价函数求各仿射参数的偏导数并令其为零，求出仿射变换参数，得到MS仿射变换跟踪结果。通过SIFT对目标区和MS跟踪区进行SIFT特征提取、匹配和仿射变换参数计算，得到SIFT仿射变换跟踪结果。最终，采用线性加权的方法融合MS和SIFT跟踪结果，获得最终的跟踪结果。研究表明，提出的算法有效地解决了跟踪矩形可自适应调整姿态的问题。
Video target tracking is one of the core technologies of the computer vision, itincorporates image processing, automatic control and artificial intelligence, etc, and hasimportant scientific theoretical significance and engineering application value in the fieldsuch as video surveillance, medical diagnosis, biological research, human-computerinteraction, autopilot and robot, etc. This thesis using Mean Shift (MS) as the core algorithmof target tracking, improving the accuracy, robustness and adaptability of target tracking asthe goal, aims to resolve the problems such as partial occlusion, deformation, illuminationvariation, scale variation and tracking rectangle with adjust attitude in video tracking.
Traditional color histogram MS algorithm only considers color statistical informationof the object, and doesn’t contain space information, so when the object color closes to thebackground color, the traditional MS algorithm easily causes tracking inaccurately or lost.Aiming at this problem, this thesis proposes a novel MS tracking algorithm with adaptiveblock color histogram, which determines block method by the size of the lastest enclosingrectangle and determines their weight coefficient by the Bhattacharyya coefficient of allblocks. Among them, the adaptive block color histogram contains adaptive block methodand spatial information of the object, and the weighted Bhattacharyya coefficient considersthe influence of different blocks to the overall similarity. Research shows that the proposedmethod can dynamically adjust tracking rectangle scale of the horizontal and verticaldirections, adaptively determining the block method of the object, and has better trackperformance than the traditional MS algorithm and fixed block MS algorithm under somecases such as partial occlusion and deformation, etc.
Considering the traditional MS algorithm is vulnerable to the influence of factors suchas illumination change, partial occlusion, image blurring, etc, this thesis introduces theimage local feature points. It presents a comparison of the most popular feature descriptorssuch as BRIEF, LAZY, ORB, RIFF, SIFT and SURF from the six aspects: rotationinvariance, scale invariance, illumination invariance, resisting vagueness, tracking accuracy and extraction time, combining the research content, and finally chooses SIFT as the featuredescriptor.
Aiming at the problem of traditional MS algorithm is vulnerable to the influence offactors such as illumination change, partial occlusion, image blurring etc, the thesisproposes a novel object tracking algorithm based on improved MS and SIFT, which iscomposed of proposed MS (initial location) and SIFT feature extraction, matching andtracking (SIFT tracking). The former is the chapter III algorithm in the thesis, and on thebasis of the former tracking results, the latter utilizes SIFT tracking method to obtain SIFTtracking results. Finally, the algorithm utilizes linear weighted method to fuse the trackingresults of improved MS and SIFT tracking, obtaining the final tracking results. Researchshows that the proposed method not only keeps the advantages of chapter III algorithm, butalso further solves the tracking problems under the situation of partial occlusion,illumination change, image blurring, etc.
In order to solve the problem which target tracking rectangle can adaptively adjustattitude, the thesis proposes a novel object tracking algorithm fused by MS and SIFT withaffine transformation. The former introduces affine transformation into the target candidatemodel, and translates complex motion tracking into the optimization problem of the costfunction. By calculating the first derivative of the cost function with respect to affineparameters and setting them to be zero, it can find the affine transformation parameters, andget MS affine transformation tracking results. On the basis of the former tracking results, thelatter utilizes the method of SIFT feature extraction, matching and tracking to obtain SIFTaffine tracking results. Finally, the algorithm utilizes linear weighted method to fuse thetracking results of MS and SIFT affine tracking, obtaining the final tracking results. Researchshows that the proposed algorithm extends the chapter III and V algorithms, and canadaptively adjust the attitude of the tracking rectangle.
传统颜色直方图的MS算法只考虑了目标颜色的统计信息，不包含目标的空间信息，当目标颜色与背景颜色相近时，容易导致不准确跟踪或跟踪丢失。针对该问题，提出了一种自适应空间颜色直方图的MS跟踪算法。该算法根据目标对象的最新外接矩形尺寸，确定对象分块方法，根据各块的Bhattacharyya系数值，确定各块的权重系数。其中，自适应分块的颜色直方图包含了自适应分块方法和目标的空间信息；加权Bhattacharyya系数考虑到不同块对整体相似度的不同影响。研究表明，提出的算法可以动态地调整跟踪矩形水平和垂直方向的尺度，自适应地确定对象的分块方法，在部分遮挡和形变等情况下，比传统的MS算法和固定分块的MS算法具有更好的跟踪性能。
考虑到传统MS算法易受光照变化、部分遮挡、图像模糊等因素的影响，引入了图像局部特征点。采用实验对比的方法从旋转不变性、尺度不变性、光照不变性、抗模糊性、跟踪精度及提取时间等六个方面，对目前最流行的六个特征点算法及描述子(BRIEF、LAZY、ORB、RIFF、SIFT和SURF)等进行比较，最终确定了SIFT作为本文采用的特征点及描述子。
针对传统MS算法易受光照变化、部分遮挡和图像模糊等因素影响的问题，提出了一种融合改进MS和SIFT的跟踪算法，该方法由改进的MS跟踪算法(初定位)和SIFT特征提取、匹配和跟踪(SIFT跟踪)组成。前者即第三章算法，后者在前者跟踪结果的基础上，采用SIFT跟踪方法得到SIFT跟踪结果。最后，通过线性加权的方法融合改进MS和SIFT的跟踪结果，获得最终的跟踪结果。研究表明，提出的算法进一步地解决了部分遮挡、光照变化和图像模糊情况下的目标跟踪问题。
为了解决目标对象跟踪矩形可自适应调整姿态的问题，提出了一种融合MS和SIFT的仿射变换目标跟踪算法。将仿射变换引入到目标的候选模型中，将复杂运动跟踪问题转化为代价函数的优化问题。通过对代价函数求各仿射参数的偏导数并令其为零，求出仿射变换参数，得到MS仿射变换跟踪结果。通过SIFT对目标区和MS跟踪区进行SIFT特征提取、匹配和仿射变换参数计算，得到SIFT仿射变换跟踪结果。最终，采用线性加权的方法融合MS和SIFT跟踪结果，获得最终的跟踪结果。研究表明，提出的算法有效地解决了跟踪矩形可自适应调整姿态的问题。
Video target tracking is one of the core technologies of the computer vision, itincorporates image processing, automatic control and artificial intelligence, etc, and hasimportant scientific theoretical significance and engineering application value in the fieldsuch as video surveillance, medical diagnosis, biological research, human-computerinteraction, autopilot and robot, etc. This thesis using Mean Shift (MS) as the core algorithmof target tracking, improving the accuracy, robustness and adaptability of target tracking asthe goal, aims to resolve the problems such as partial occlusion, deformation, illuminationvariation, scale variation and tracking rectangle with adjust attitude in video tracking.
Traditional color histogram MS algorithm only considers color statistical informationof the object, and doesn’t contain space information, so when the object color closes to thebackground color, the traditional MS algorithm easily causes tracking inaccurately or lost.Aiming at this problem, this thesis proposes a novel MS tracking algorithm with adaptiveblock color histogram, which determines block method by the size of the lastest enclosingrectangle and determines their weight coefficient by the Bhattacharyya coefficient of allblocks. Among them, the adaptive block color histogram contains adaptive block methodand spatial information of the object, and the weighted Bhattacharyya coefficient considersthe influence of different blocks to the overall similarity. Research shows that the proposedmethod can dynamically adjust tracking rectangle scale of the horizontal and verticaldirections, adaptively determining the block method of the object, and has better trackperformance than the traditional MS algorithm and fixed block MS algorithm under somecases such as partial occlusion and deformation, etc.
Considering the traditional MS algorithm is vulnerable to the influence of factors suchas illumination change, partial occlusion, image blurring, etc, this thesis introduces theimage local feature points. It presents a comparison of the most popular feature descriptorssuch as BRIEF, LAZY, ORB, RIFF, SIFT and SURF from the six aspects: rotationinvariance, scale invariance, illumination invariance, resisting vagueness, tracking accuracy and extraction time, combining the research content, and finally chooses SIFT as the featuredescriptor.
Aiming at the problem of traditional MS algorithm is vulnerable to the influence offactors such as illumination change, partial occlusion, image blurring etc, the thesisproposes a novel object tracking algorithm based on improved MS and SIFT, which iscomposed of proposed MS (initial location) and SIFT feature extraction, matching andtracking (SIFT tracking). The former is the chapter III algorithm in the thesis, and on thebasis of the former tracking results, the latter utilizes SIFT tracking method to obtain SIFTtracking results. Finally, the algorithm utilizes linear weighted method to fuse the trackingresults of improved MS and SIFT tracking, obtaining the final tracking results. Researchshows that the proposed method not only keeps the advantages of chapter III algorithm, butalso further solves the tracking problems under the situation of partial occlusion,illumination change, image blurring, etc.
In order to solve the problem which target tracking rectangle can adaptively adjustattitude, the thesis proposes a novel object tracking algorithm fused by MS and SIFT withaffine transformation. The former introduces affine transformation into the target candidatemodel, and translates complex motion tracking into the optimization problem of the costfunction. By calculating the first derivative of the cost function with respect to affineparameters and setting them to be zero, it can find the affine transformation parameters, andget MS affine transformation tracking results. On the basis of the former tracking results, thelatter utilizes the method of SIFT feature extraction, matching and tracking to obtain SIFTaffine tracking results. Finally, the algorithm utilizes linear weighted method to fuse thetracking results of MS and SIFT affine tracking, obtaining the final tracking results. Researchshows that the proposed algorithm extends the chapter III and V algorithms, and canadaptively adjust the attitude of the tracking rectangle.
引文
[1] Collies R T, et al. A System for Video Surveillance and Monitoring: VSAM FinalReport[R]. Techinical Report CMU-RI-TR-04-12, Carnegie Mellon University,2000.
[2] Haritaoglu I, Harwood D, Davis LS. W4: Who? When? Where? What? A Real TimeSystem for Detecting and Tracking People[A]. Proceeding of the InternationalConference on Automatic Face and Gesture Recognition[C]. New York: IEEE Press,1998,222-227.
[3] Haritaoglu I, Harwood D, Davis LS. W4: Real-time Surveillance of People and TheirActivities[J]. IEEE Trans. On Pattern Analysis and Machine Intellience.2000,22(8):809-830.
[4] Prange J D. Detecting, Recognizing and Understanding Video Events in SurveillanceVideo[A]. Proceeding of the International Conference on Advanced Video and SignalBased Surveillance(AVSS’03)[C]. New York: IEEE Press,2003:4.
[5] I. Charlton, P. Tate, P. Smyth, and L. Roren. Repeatability of an Optimised Lower BodyModel[J]. Gait&Posture,2004,20(2):213–221.
[6] P. Mountney and G.-Z. Yang. Soft Tissue Tracking For Minimally Invasive Surgery:Learning Local Deformation Online[A]. Proceedings of the International Conferenceon Medical Image Computing and Computer-Assisted Intervention[C], Berlin,Germany,2008,364-572.
[7] D.A. Leung, J.F. Debatin, S. Wildermuth, N. Heske, C.L. Dumoulin, R.D. Darrow, M.Hauser, C.P. Davis, and GK Von Schulthess. Real-time Biplanar Needle Tracking forInterventional mr Imaging Procedures[J]. Radiology,1995,197(2):485–488.
[8] E. Kochavi, D. Goldsher and H. Azhari. Method for Rapid MRI Needle Tracking[J].Magnetic Resonance in Medicine,2004,51(5):1083-1087.
[9] K. Mori, D. Deguchi, K. Akiyama, T. Kitasaka, C.R. Maurer Jr., Y. Suenaga, H.Takabatake, M. Mori and H. Natori. Hybrid Bronchoscope Tracking Using a MagneticTracking Sensor and Image Registration[A]. Proceedings of the InternationalConference on Medical Image Computing and Computer-Assisted Intervention[C],Palm Springs CA,2005,543-550.
[10] E. Maggio, A. Cavallaro. Video Tracking Theory and Practice[M]. A John Wiley andSons, Ltd., Publication,2011.
[11] Cipolla R, Pentland A. Computer Vision for Human-Machine Interaction[M].Cambridge: Cambridge University Press,1998.
[12] Row S, Blake A. Statistical Mosaics for Tracking[J]. Image and Vision Computing,1996,14(3):549-564.
[13] Wellner P. Interacting with Paper on the Digital Desk[J]. Communications of the ACM,1993,36(7):87-96.
[14]徐一华，李京峰等.人体三维运动实时跟踪与建模系统[J].自动化学报,2006,32(4):560-567.
[15] Oka K, Sato Y, Koike H. Real-Time Fingertip Tracking and Gesture Recognition[J].IEEE Computer Graphics and Applications,2002,22(6):64-71.
[16]王西颖,张习文,戴国忠.一种面向实时交互的变形手势跟踪方法[J].软件学报,2007,18(10):2423-2433.
[17] Bradski G R, Clara S. Computer Vision Face Tracking for Use in a Perceptual UserInterface[J]. Intelligence Technology Journal,1998,2(2):1-15.
[18] Dedeo G, Kanada T, S Baker. The Asymmetry of Image Registration and itsApplication to Face Tracking[J]. IEEE Trans. on Pattern Analysis and MachineIntelligence,2007,29(5):807-823.
[19] Maggioni C, Rottge H. Virtual Touchscreen-A Novel User Interface Made ofLight-principles, Metaphors and Experiences[A]. Proceedings of the8th InternationalConference on Human Computer Interaction[C]. New York: IEEE Press,1999:301-305.
[20] Cipolla R, Hollinghurst N J. Human-robot Interface by Pointing with UncalibratedStereo Vision[J]. Image and Vision Computing,1996,14(3):171-178.
[21] Maccormick J, Blake A. Partitioned Sampling, Articulated Objects and Interface-qulityHand Tracking[A]. Proceedings of the7th European Conference on ComputerVision[C]. Berlin: Springer-Verlag,2000,2:3-19.
[22] Isard M, Blake A. A Mixed-state Condensation Tracker with Automatic ModelSwitching[A]. Proceedings of the6th International Conference on Computer Vision[C].New York: IEEE Press,1998:107-112.
[23] Desouza G N, Kak A C. Vision for Mobile Robot Navigation: A Survey[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,24(2):237-267.
[24] Honda worldwide P3. http://world.honda.com/ASIMO/P3/
[25] Jung B, Sukhatme G S. Detecting Moving Objects Using a Single Camera on a MobileRobot in an Outdoor Environment[A]. Proceedings of International Conference onIntelligent Autonomous Systems[C]. Amsterdam, Netherlands,2004:980-987.
[26] Jung B, Sukhatme G. Tracking Targets using Multiple Robots: the Effect ofEnvironment Occlusion[J]. Autonomous Robots Journal,2002,13(3):191-205.
[27] Treptow A, Masselli A, Zell A. Real-time Object Tracking for Soccer-robots withoutColor Information[J]. Robotics and Autonomous Systems,2004,48(1):41-48.
[28] Liu P R, Meng M Q H, Liu P X. Moving Object Segmentation and Detection forMonocular Robot Based on Active Contour Model[J]. ELECTRONICS LETTERS2005,41(24):1320-1322.
[29] Montesano L, MontanoIntl L. Identification of Moving by a Team of Robots Based onKinematic Information[A]. Proceedings of Conference on Intelligent Robots andSystems[C]. New York: IEEE Press,2003,1:284-290.
[30]王璐,崔益安,苏虹等.移动机器人的运动目标实时检测与跟踪[J].计算机工程与应用,2005,41(15):30-33.
[31]李培华.序列图像中运动目标跟踪方法[M].北京：科学出版社，2010.
[32] Terzopoulos D, Szeliski R. Tracking with Kalman Snakes[J]. Blake A, Yuille A. ActiveVision. Cambridge: MIT Press, Cambridge, MA,1992:3-20.
[33] Blake A, Curwen R, Zisserman A. A Framework for Spatio-temporal Control in theTracking of Visual Contours[J]. International Journal Computer Vision,1993,11(2):127-145.
[34] Peterfreund N. Robust Tracking with Spatio-velocity Snakes Kalman FilteringApproach[A]. International Conference Computer Vision[C], Bombay, India,1998:433-439.
[35] Gordon N J, Salmond D J, Smith A F M. Novel Approach to Nonlinear/Non-GaussianBayesian State Estimation[A]. IEEE Proceedings-F[C],1993,140(2):107-113.
[36] Kitagawa G. Monte Carlo Filter and Smoother for Non-Gaussian Nonlinear StateSpace Models[J]. Computational and Graphical Statistics,1996,5:1-25.
[37] Doucet A, Godsill S J, Andrieu C. On Sequential Monte Carlo Sampling Methods forBayesian Filtering[J]. Statistics and Computing,2000,10(3):197-208.
[38] Isard M, Blake A. Contour Tracking by Stochastic Propagation of ConditionalDensity[A]. Proceedings of European Conference on Computer Vision[C]. Berlin:Springer-Verlag,1996:343-356.
[39] Perez P, Hue C, Vermaak J, et al. Color-based Probabilistic Tracking[A]. Proceedingsof European Conference on Computer Vision[C]. Berlin: Springer-Verlag,2002:661-675.
[40] Nummiaro K, Koller-Meier E, Gool L V. An Adaptive Color-based Particle Filter[J].Image and Vision Computing,2003,21(1):100-110.
[41] Wu Y, Huang T S. Robust Visual Tracking by Integrating Multiple Cues Based onCo-inference Learning[J]. International Journal of Computer Vision,2004,58(1):55-71.
[42] Jacquot A, Sturm P, Ruch O. Adaptive Tracking of Non-rigid Objects Based on ColorHistograms and Automatic Parameter Selection[J]. IEEE Workshop on Motion andVideo Computing, New York: IEEE Press,2005:103-109.
[43] Stauffer C, Grimson W. Learning Patterns of Activity Using Real-time Tracking[J].IEEE Trans. Pattern Anal. Machine Intelligence,2000,22(8):747-757.
[44] McKenna S J, Gong S, Raja Y. Modelling Facial Color and Identity with GaussianMixtures[J]. Pattern Recognition.1998,31(12):1883-1892.
[45] Isard M, Blake A. Icondensation: Unifying Low-level and High-level Tracking in aStochastic Framework[A]. Proceedings of European Conference on ComputerVision[C]. Berlin: Springer-Verlag,1998:893-908.
[46] Spengler M, Schiele B. Towards Robust Multi-cue Integration for Visual Tracking[J].Machine Vision and Applications,2003,14(1):50-58.
[47] Okuma K, Taleghani A, Nando de Freitas, et al. A Boosted Particle Filter: MultitargetDetection and Tracking[A]. Proceedings of European Conference on ComputerVision[C]. Berlin: Springer-Verlag,2004:28-39.
[48] Yang C, Durraiswami R, Davis L S. Fast Multiple Object Tracking Via a HierarchicalParticle Filter[A]. International Conference on Computer Vision[C], New York: IEEEPress,2005:212-219
[49] Viola P, Jones M J. Rapid M J. Rapid Object Detection Using a Boosted Cascade ofSimple Features[A]. IEEE Conference on Computer Vision and Pattern Recognition[C],New York: IEEE Press,2001:511-518.
[50] Jones M, Viola P. Fast Multi-view Face Detection[A]. IEEE Conference on ComputerVision and Pattern Recognition[C]. New York: IEEE Press,2003:587-594.
[51] Ning H Z, Tan T N, Wang L, et al. People Tracking Based on Motion Model andMotion Constraints with Automatic Initialization[J]. Pattern Recognition,2004,37(7):1423-1440.
[52] Wang J Y, Chen X L, Gao W. Online Selecting Discriminative Tracking Features usingParticle Filter[A]. IEEE Conference on Computer Vision and Pattern Recognition[C].New York: IEEE Press,2005:1037-1042.
[53] Wang Q, Ai H Z, Xu G Y. Learning Based Tracking of Ccomplex Non-rigid Motion[J].Journal of Computer Science and Technlogy,2004,19(4):489-500.
[54] Li P H, Zhang T W, Pece A E C. Visual Contour Tracking Based on Particle Filters[J].Image and Vision Computing,2003,21(1):111-123.
[55] K. Fukunaga and L.D. Hostetler. The Estimation of the Gradient of a Density Function,with Applications in Pattern Recognition[J]. IEEE Trans. Information Theory.1975(21)：32-40.
[56] Y. Cheng, Mean Shift, Mode Seeking, and Clustering[J]. IEEE Trans. Pattern Analysisand Machine Intelligence.1995,17(8):790-799.
[57] Comaniciu D, Ramesh V, Meer P. Real-Time Tracking of Non-rigid Objects UsingMean Shift[A]. Proceedings of the IEEE Conference on Computer Vision and PatternRecognition,2000:142-149.
[58] Comaniciu D, Ramesh V, Meer P. Kernel-Based Object Tracking[J]. IEEE Trans. OnPattern Analysis and Machine Intelligence,2003,25(5):564-575.
[59] Lindeberg T. Feature Detection with Automatic Scale Selection[J]. InternationalJournal of Computer Vision,1998,30(2):194-203.
[60] Collins R T. Mean Shift Blob Tracking through Scale Space[A]. IEEE Conference onComputer Vision and Pattern Recognition[C], New York: IEEE Press,2003:234-240.
[61]李乡儒，吴福朝，胡占义.均值漂移算法的收敛性[J].软件学报，2005,16(3):365-374.
[62] Li X R, Hu Z Y, Wu F C. A Note on the Convergence of the Mean Shift[J]. PatternRecognition,2007,40(6):1756-1762.
[63] Zhao Q, Tao H. Object Tracking Using Color Correlogram[A]. Proceedings of IEEEWorkshop on Visual Survellance and Performance Evaluation of Tracking andSurveillance[C], New York: IEEE Press,2005:263-270.
[64] Birchfield S T, Rangarajan S. Spatiograms Versus Histograms for Region-basedTracking[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C], New York: IEEE Press,2005:1158-1163.
[65]李培华.一种改进的MeanShift跟踪算法[J].自动化学报，2007，33(4):347-354.
[66]杜凯，巨永锋，靳引利，李刚.自适应空间颜色直方图的Meanshift跟踪算法[J].武汉理工大学学报，2012,34(6):140-144.
[67] Liu T L, Chen H T. Real-time Tracking Using Trust-region Methods[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2004,26(3):397-402.
[68] Yang C J, Duraiswami R, de Menthon D, et al. Mean-shift Analysis UsingQuasi-newton Methods[A]. Proceedings of IEEE Conference on Image Processing[C].New York: IEEE Press,2003:447-450.
[69] Fashing M, Tomasi C. Mean Shift is a Bound Optimization[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,2005,27(3):471-474.
[70] Bajramovic F, Grabl C, Denzler J. Efficient Combination of Histograms for Real-timeTracking Using Mean-shift and Trust-region Optimization[A]. Proceedings of27thDAGM Symposium on Pattern Recognition[C], Berlin: Springer-Verlag,2005:254-261.
[71] Zhang H, Huang Z, Huang W, et al. Kernel-based Method for Tracking Objects withRotation and Translation[A]. Proceedings of International Conference on PatternRecognition[C], New York: IEEE Press,2004:728-731.
[72] Zhang H H, Huang W M, Huang Z Y, et al. Affine Object Tracking with Kernel-basedSpatial Color Representation[A]. IEEE Conference on Computer Vision and PatternRecognition[C], New York: IEEE Press,2005:293-300.
[73] Yilmaz A. Object Tracking by Asymmetric Kernel Mean Shift with Automatic Scaleand Orientation Selection[A]. Proceeding of IEEE Conference on Computer Vision andPattern Recognition[C], New York: IEEE Press,2007:1-6.
[74] Elgammal A, Duraiswami R, Davis L S. Probabilistic Tracking in Joint Feature-spatialSpaces[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C], New York: IEEE Press,2003:781-788.
[75] Elgammal A, Duraiswami R, Davis L S. Efficient Non-parametric Adaptive ColorModeling Using Fast Gauss Transform[A]. Proceedings of IEEE Conference onComputer Vision and Pattern Recognition[C]. New York: IEEE Press,2001:781-788.
[76] Yang C J, Duraiswami R, Davis L. Efficient Mean-shift Tracking Via a New SimilarityMeasure[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C], New York: IEEE Press,2005:176-183.
[77] Hager G, Dewan M, Stewart C. Multiple Kernel Tracking with SSD[A]. Proceedingsof IEEE Conference on Computer Vision and Pattern Recognition[C]. New York: IEEEPress,2004:790-797.
[78] Leichter I, Lindenbaum M, Rivlin E. Visual Tracking by Affine Kernel Fitting UsingColor and Object Boundary[A]. Proceedings of International Conference on ComputerVision[C]. New York: IEEE Press,2007:1-6.
[79] Zivkovic Z, Krose B. An EM-like Algorithm for Color-Histogram-Based ObjectTracking[A]. Proceedings of IEEE Conference on Computer Vision, and PatternRecognition[C], New York: IEEE Press,2004:798-803.
[80] Shen C H, Brooks M J, Hegnel A. Fast Global Kernel Density Mode Seeking:Applications to Localization and Tracking[J]. IEEE Transaction on Image Processing,2007,16(5):1457-1469.
[81] Zhao Q, Brennan S, Tao H, et al. Differential EMD tracking[A]. Proceedings ofInternational Conference on Computer Vision[C]. New York: IEEE Press,2007:1-8.
[82] Han B, Comaniciu D, Zhu Y, et al. Sequential Kernel Density Approximation and itsApplication to Real-time Visual Tracking[J]. IEEE Transaction on Pattern Analysis andMachine Intelligence,2008,30(7):1186-1197.
[83] Park M, Liu Y, Collins R. Efficient Mean Shift Belief Propagation for VisionTracking[A]. Proceedings of IEEE Conference Computer Vision and PatternRecognition[C]. New York: IEEE Press,2008:1-8.
[84] Fan Z, Yang M, Wu Y. Multiple Collaborative Kernel Tracking[J]. IEEE Transactionon Pattern Analysis and Machine Intelligence,2007,29(7):1268-1273.
[85] Kass M, Witkings A, Terzopoulos. Snakes: Active Contour Models[J]. InternationalJournal of Computer Vision,1988,1(4):321-331.
[86] Aubert G, Kornprobst P. Mathematical Problems in Image Processing: PartialDifferential Equations and the Calculus of Variations[M]. Berlin: Springer-Verlag,2006.
[87] Caselles V, Kimmel R, Sapiro G. Geodesic active contours[J]. International Journal ofComputer Vision,1997,22(3):61-79.
[88] Paragios N, Deriche R. Geodesic Active Regions and Level Set Methods for MotionEstimation and Tracking[J]. Computer Vision and Image Understanding,2005,97(3):259-282.
[89] Paragios N, Deriche R. Geodesic Active Contours and Level Set for the Detection andTracking of Moving Objects[J]. IEEE Trans. Pattern Anal. Machine Intell.,2000,22(3):266-280.
[90] Boltz S, Herbulot A, Debreuve E, et al. Motion and Appearance Nonparametric JointEntropy for Video Segmentation[J]. International Journal of Computer Vision,2008,80(2):242-259.
[91] Jehan-Besson S, Barlaud M, Aubert G, et al. Shape Gradients for HistogramSegmentation Using Active Contours[A]. International Conference on ComputerVision[C]. New York: IEEE Press,2003:408-415.
[92] Freedman D, Zhang T. Active Contours for Tracking Distributions[J]. IEEE Trans.Image Processing,2004,13(4):518-526.
[93] Cremers D. Nonlinear Dynamical Shape Priors for Level Set Segmentation[A]. IEEEConference on Computer Vision and Pattern Recognition[C], New York: IEEE Press,2007:1-7.
[94] Sundaramoorthi G, Yezzi A, Mennucci A, et al. New Possibilities with Sobolev ActiveContours[J]. International Journal of Computer Vision, DOI:10.1007/s11263-008-0133-9,2009.
[95] Chan T F, Vese L A. Active Contours without Edges[J]. IEEE Trans. Image Processing,2001,10(2):266-277.
[96] Rathi Y, Vaswani N, Tannenbaum A, et al. Tracking Deforming Objects Using ParticleFiltering for Geometric Active Contours[J]. IEEE Trans. Pattern Anal. MachineIntelligence,2007,29(8):1470-1475.
[97] Cremers D, Rousson M, Deriche R. A Review of Statistical Approaches to Level SetSegmentation: Integrating Color, Texture, Motion and Shape[A]. International Journalof Computer Vision[C],2007,72(2):195-215.
[98] Aubert G, Barlaud M, Faugeras O, et al. Image Segmentation Using Active Contours:Calculus of Variations or Shape Gradients?[J]. SIAM Journal of Applied Mathematics,2003,63(6):2128-2154.
[99] Jehan-Besson S, Barlaud M, Aubert G. DREAM2S: Deformable Regions Driven by anEulerian Accurate Minimization Method for Image and Video Segmentation[J].International Journal of Computer Vision,2003,53(1):45-70.
[100] Osher S, Sethian J A. Fronts Propagating with Curvature-Dependent Speed:Algorithms Based on Hamilton-Jacobi Formulations[J]. Journal of ComputationalPhysics,1988,79:12-49.
[101] Sethian J A. Level Set Methods and Fast Marching Methods[M]. Cambridge:Cambridge Universty Press,1999.
[102] Goldenberg R, Kimmel R, Rivlin E, et al. Fast Geodesic Active Contours[J]. IEEETrans. on Image Processing,2001,10(10):1467-1475.
[103] Papandreou G, Maragos P. Multigrid Geometric Aactive Contour Models[J]. IEEETrans. on Image Processing,2007,16(1):229-240.
[104]马波，张田文.基于曲线进化的目标检测算法[J].信号处理，2003,3:237-241.
[105]于慧敏，徐艺，刘继忠，等.基于水平集的多运动目标时空分割与跟踪[J].中国图像图形学报，2007,7:1218-1223.
[106]周则明，王元全，王平安，等.基于水平集的3D左心室表面重建[J].计算机研究与发展,2005,42(7):1173-1178.
[107]陈卫刚，戚飞虎.一种改进的窄带算法在彩色序列图像人脸跟踪中的应用[J].电子与信息学报，2005,4:540-543.
[108]黄福珍，苏剑波.基于Level Set方法的人脸轮廓提取与跟踪[J].计算机学报，2003,26(4):491-496.
[109]王元全，贾云得.一种新的心脏核磁共振图像分割方法[J].计算机学报，2007,30(1):129-136.
[110]查宇飞，张育，毕笃彦.基于区域活动轮廓运动目标跟踪方法研究[J].中国图像图形学报，2006,12:1844-1848.
[111]罗嘉，韦志辉.基于几何活动轮廓模型的目标跟踪与快速运动估计[J].中国图像图形学报，2009,14(7):1361-1368.
[112]周雪，胡卫明.融合颜色和增量形状先验的目标轮廓跟踪[J].自动化学报,2009,35(11):1394-1402.
[113] Li P H, Xiao L J. Histogram-Based Partial Differential Equation for ObjectTracking[A]. The7th International Conference on Advances in Pattern Recognition,ICAPR09[C]. New York: IEEE Press,2009:286-289.
[114] Jepson A. D., Fleet D, J., El-Maraghi T. F. Robust Online Appearance Models forVisual Tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(10):415-522.
[115] Ross D., Lim J., Yang M.-H. Adaptive Probabilistic Visual Tracking with IncrementalSubpace Update[A]. Proceedings of the eighth European conference on computervision[C]. Berlin: Springer,2004.
[116] Wu Y., Huang T. S. Robust Visual Tracking by Integrating Multiple Cues Based onCo-Inference Learning[J]. International Journal of Computer Vision.2004,58(1):55-71.
[117] Cannons K., Wildes R. Spatiotemporal Oriented Energy Features for VisualTracking[A]. In: Asian Conference on Computer Vision[C].2007. pp532-543.
[118] Senior A., Hampapur A., Tian Y.-L., et al. Appearance Models for OcclusionHandling[J]. Image and Vision Computing,2006,24(11):1233-1243.
[119] Yilmaz A., Xin L., Shah M. Contour-based Object Tracking with Occlusion Handlingin Video Acquired Using Mobile Cameras[J]. IEEE Transcations on Pattern Analysisand Machine Intelligence,2004,26(11):1531-1536.
[120] Tao Y., Quan P., Jing L., et al. Real-time Multiple Objects Tracking with OcclusionHandling in Dynamic Scenes[A]. Proceedings of IEEE Conference on Computer Visionand Pattern Recognition[C]. San Diego, CA, USA: IEEE,2005:970-975.
[121] Onoguchi K. Shadow Elimination Method for Moving Object Detection[A].Proceedings of IEEE International Conference on Pattern Recognition[C]. New York:IEEE Press,1998,1:583-587.
[122] Comaniciu, D. and P. Meer. Mean Shift Analysis and Application[A]. Proceedings ofthe7th IEEE International Conference on Computer Vision[C]. Kerkyra, Greece,1999,pp1197-1203.
[123] Comaniciu, D. and P. Meer. Mean Shift: A Robust Application toward Feature SpaceAnalysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603-619.
[124] Comaniciu, D. and P. Meer. Robust Analysis of Feature Spaces: Color ImageSegmentation[A]. Proceedings of IEEE Conference Computer Vision and PatternRecognition[C]. San. Juan, Puerto Rico1997:750–755.
[125] Comaniciu, D., V. Ramesh, and P. Meer. The Variable Bandwidth Mean Shift andData-driven Scale Selection[A]. Proceedings of the8th International Conference onComputer Vision[C]. Vancouver, Canada,2001:438-445.
[126] Yang, C,. R. Duraiswami, and L. Davis. Similarity Meansure for NonparametricKernel Density Based Object Tracking[A]. Eighteenth Annual Conference on NeuralInformation Processing Systems[C]. Victoria, British Columbia, Canada: NIPS,2004.
[127] Collins, R.T. Mean-Shift Blob Tracking through Scale Space[A]. in IEEE Conferenceon Computer Vision and Pattern Recognition[C].2003, Madison, Wisconsin,2003:234–240.
[128] Nummiaro, K., E. Koller-Meier, and L. Van Goo. Color Features for TrackingNon-Rigid Objects[J]. Chinese Journal of Automation,2003,29(3):345-355.
[129] McFarlane, N.J.B. and C.P. Schofield. Segmentation and Tracking of Piglets inImages [J]. Machine Vision and Applications,1995,8:187-193.
[130] Kailath, T., The Divergence and Bhattacharyya Distance Measures in SignalSelection [J]. IEEE Trans. Comm. Technology.,1967.15:52-60.
[131] Swain, M.J. and D.H. Ballard, Color Indexing[J]. International Journal of ComputerVision,1991.7(1):11-32.
[132] Maggio, E. and A. Cavallaro. Hybrid Particle Filter and Mean Shift Tracker withAdaptive Transition Model[A]. In Acoustics, Speech, and Signal Processing,2005.Proceedings[C].2005.
[133]袁亚湘，孙文瑜.最优化理论与方法[M].北京：科学出版社，2001.
[134] David K, Ward C. Numberical Analysis: Mathematics of Scientific Computing[M].Washington: American Mathematical Society,2002.
[135] Nocedal J, Wright S J. Numerical Optimization[M]. New York：Springer-Verlag,1999.
[136] D. Xu, Y. Wang, and J. An. Applying a New Spatial Color Histogram in Mean-ShiftBased Tracking Algorithm[A]. Image and Vision Computing[C]. New Zealand,2005.
[137] E. Maggio, A. Cavallaro. Multi-Part Target Representation for Color Tracking[A].Proc. IEEE International Conference on Image Processing[C],2005:729-732.
[138]袁霄,王丽萍.基于MeanShift算法的运动人体跟踪[J].计算机工程与科学,2008,30(4):46-49.
[139]张恒,李由,李立春,于起峰.基于显著性加权的Mean Shift跟踪方法[J].光学技术,2008,34(3):404-407.
[140]胡铟,杨静宇.基于分块颜色直方图的MeanShift跟踪算法[J].系统仿真学报,2009,21(10):2936-2955.
[141]王东峰，张丽飞，刘小军，邹谋炎.基于广义特征点匹配的全自动图像配准[J].电子与信息学报，2005，27(7):1013-1016.
[142]王永明，王贵锦.图像局部不变性特征与描述[M].北京：国防工业出版社，2010.
[143] Lindeberg T. Detecting Salient Blob-Like Image Structures and Their Scales with aScale-Space Primal Sketch: A Method for Focus-of-Attention[J]. International Journalof Computer Vision,1993,11(3).
[144] Lindeberg T. Feature Detection with Automatic Scale Selection[J]. InternationalJournal of Computer Vision,1998,30(2).
[145] Lowe D G. Distinctive Image Features from Scale-invariant Keypoints[J].International Journal of Computer Vision,2004,60(2).
[146] Bay H, Tuytelaars T, Van Gool L. SURF: Speeded up Robust Features[C]. In ECCV,2006.
[147] Bay H, Tuytelaars T, Van Gool L. Speeded-Up Robust Features(SURF)[J]. ComputerVision and Image Understanding,2008.
[148] Bay H. From Wide-baseline Point and Line Correspondences to3D[D]. Ph.D.Thesis,ETH Zurich,2006.
[149] Bay H, Fasel B, Van Gool L. Interactive Museum Guide: Fast and RobustRecognition of Museum Objects[A]. Proceedings of the First International Workshopon Mobile Vision,2006.
[150] Viola P, Jones M. Rapid Object Detection Using a Boosted Cascade of simplefeatures[C]. In IEEE Conference on Computer Vision and Pattern Recoghition,2001,1.
[151] Simard P, Bottou L, Haffner P, et.al. Boxlets: a Fast Convolution Algorithm forSignal Processing and Neural Networks[J]. Advances in Neural Information ProcessingSystems,1999,11.
[152] K. Eugene. Feature Descriptor Comparison Report[EB/OL]. http://computer-vision-talks.com/2011/08/feature-descriptor-comparison-report/,2011.
[153] M. Calonder, V. Lepetit, C. Strecha, P. Fua. BRIEF: Binary Robust IndependentElementary Features[C]. In ECCV2010.
[154] E. Rublee, V. Rabaud, K. Konolige, G. Bradski. ORB: an Efficient Alternative toSIFT or SURF[C]. In ICCV2011.
[155] Zhou H Y, Yuan Y, and Shi C. Object Tracking Using SIFT Features and MeanShift[J]. Computer Vision and Image Understanding,2009,345-352.
[156]李培华,肖莉娟.基于Mean Shift的相似性变换和仿射变换目标跟踪算法[J].中国图象图形学报,2011,16(2):258-266.
[157]张贤达.矩阵分析与应用[M].北京：清华大学出版社,2004.
[158] Heikkila M, Pietikainen M. A Texture-based Method for Modelling the Backgroundand Detecting Moving Objects[J]. IEEE Transaction on Pattern Analysis and MachineIntelligence,2006,28(4):657-662.
[2] Haritaoglu I, Harwood D, Davis LS. W4: Who? When? Where? What? A Real TimeSystem for Detecting and Tracking People[A]. Proceeding of the InternationalConference on Automatic Face and Gesture Recognition[C]. New York: IEEE Press,1998,222-227.
[3] Haritaoglu I, Harwood D, Davis LS. W4: Real-time Surveillance of People and TheirActivities[J]. IEEE Trans. On Pattern Analysis and Machine Intellience.2000,22(8):809-830.
[4] Prange J D. Detecting, Recognizing and Understanding Video Events in SurveillanceVideo[A]. Proceeding of the International Conference on Advanced Video and SignalBased Surveillance(AVSS’03)[C]. New York: IEEE Press,2003:4.
[5] I. Charlton, P. Tate, P. Smyth, and L. Roren. Repeatability of an Optimised Lower BodyModel[J]. Gait&Posture,2004,20(2):213–221.
[6] P. Mountney and G.-Z. Yang. Soft Tissue Tracking For Minimally Invasive Surgery:Learning Local Deformation Online[A]. Proceedings of the International Conferenceon Medical Image Computing and Computer-Assisted Intervention[C], Berlin,Germany,2008,364-572.
[7] D.A. Leung, J.F. Debatin, S. Wildermuth, N. Heske, C.L. Dumoulin, R.D. Darrow, M.Hauser, C.P. Davis, and GK Von Schulthess. Real-time Biplanar Needle Tracking forInterventional mr Imaging Procedures[J]. Radiology,1995,197(2):485–488.
[8] E. Kochavi, D. Goldsher and H. Azhari. Method for Rapid MRI Needle Tracking[J].Magnetic Resonance in Medicine,2004,51(5):1083-1087.
[9] K. Mori, D. Deguchi, K. Akiyama, T. Kitasaka, C.R. Maurer Jr., Y. Suenaga, H.Takabatake, M. Mori and H. Natori. Hybrid Bronchoscope Tracking Using a MagneticTracking Sensor and Image Registration[A]. Proceedings of the InternationalConference on Medical Image Computing and Computer-Assisted Intervention[C],Palm Springs CA,2005,543-550.
[10] E. Maggio, A. Cavallaro. Video Tracking Theory and Practice[M]. A John Wiley andSons, Ltd., Publication,2011.
[11] Cipolla R, Pentland A. Computer Vision for Human-Machine Interaction[M].Cambridge: Cambridge University Press,1998.
[12] Row S, Blake A. Statistical Mosaics for Tracking[J]. Image and Vision Computing,1996,14(3):549-564.
[13] Wellner P. Interacting with Paper on the Digital Desk[J]. Communications of the ACM,1993,36(7):87-96.
[14]徐一华，李京峰等.人体三维运动实时跟踪与建模系统[J].自动化学报,2006,32(4):560-567.
[15] Oka K, Sato Y, Koike H. Real-Time Fingertip Tracking and Gesture Recognition[J].IEEE Computer Graphics and Applications,2002,22(6):64-71.
[16]王西颖,张习文,戴国忠.一种面向实时交互的变形手势跟踪方法[J].软件学报,2007,18(10):2423-2433.
[17] Bradski G R, Clara S. Computer Vision Face Tracking for Use in a Perceptual UserInterface[J]. Intelligence Technology Journal,1998,2(2):1-15.
[18] Dedeo G, Kanada T, S Baker. The Asymmetry of Image Registration and itsApplication to Face Tracking[J]. IEEE Trans. on Pattern Analysis and MachineIntelligence,2007,29(5):807-823.
[19] Maggioni C, Rottge H. Virtual Touchscreen-A Novel User Interface Made ofLight-principles, Metaphors and Experiences[A]. Proceedings of the8th InternationalConference on Human Computer Interaction[C]. New York: IEEE Press,1999:301-305.
[20] Cipolla R, Hollinghurst N J. Human-robot Interface by Pointing with UncalibratedStereo Vision[J]. Image and Vision Computing,1996,14(3):171-178.
[21] Maccormick J, Blake A. Partitioned Sampling, Articulated Objects and Interface-qulityHand Tracking[A]. Proceedings of the7th European Conference on ComputerVision[C]. Berlin: Springer-Verlag,2000,2:3-19.
[22] Isard M, Blake A. A Mixed-state Condensation Tracker with Automatic ModelSwitching[A]. Proceedings of the6th International Conference on Computer Vision[C].New York: IEEE Press,1998:107-112.
[23] Desouza G N, Kak A C. Vision for Mobile Robot Navigation: A Survey[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,24(2):237-267.
[24] Honda worldwide P3. http://world.honda.com/ASIMO/P3/
[25] Jung B, Sukhatme G S. Detecting Moving Objects Using a Single Camera on a MobileRobot in an Outdoor Environment[A]. Proceedings of International Conference onIntelligent Autonomous Systems[C]. Amsterdam, Netherlands,2004:980-987.
[26] Jung B, Sukhatme G. Tracking Targets using Multiple Robots: the Effect ofEnvironment Occlusion[J]. Autonomous Robots Journal,2002,13(3):191-205.
[27] Treptow A, Masselli A, Zell A. Real-time Object Tracking for Soccer-robots withoutColor Information[J]. Robotics and Autonomous Systems,2004,48(1):41-48.
[28] Liu P R, Meng M Q H, Liu P X. Moving Object Segmentation and Detection forMonocular Robot Based on Active Contour Model[J]. ELECTRONICS LETTERS2005,41(24):1320-1322.
[29] Montesano L, MontanoIntl L. Identification of Moving by a Team of Robots Based onKinematic Information[A]. Proceedings of Conference on Intelligent Robots andSystems[C]. New York: IEEE Press,2003,1:284-290.
[30]王璐,崔益安,苏虹等.移动机器人的运动目标实时检测与跟踪[J].计算机工程与应用,2005,41(15):30-33.
[31]李培华.序列图像中运动目标跟踪方法[M].北京：科学出版社，2010.
[32] Terzopoulos D, Szeliski R. Tracking with Kalman Snakes[J]. Blake A, Yuille A. ActiveVision. Cambridge: MIT Press, Cambridge, MA,1992:3-20.
[33] Blake A, Curwen R, Zisserman A. A Framework for Spatio-temporal Control in theTracking of Visual Contours[J]. International Journal Computer Vision,1993,11(2):127-145.
[34] Peterfreund N. Robust Tracking with Spatio-velocity Snakes Kalman FilteringApproach[A]. International Conference Computer Vision[C], Bombay, India,1998:433-439.
[35] Gordon N J, Salmond D J, Smith A F M. Novel Approach to Nonlinear/Non-GaussianBayesian State Estimation[A]. IEEE Proceedings-F[C],1993,140(2):107-113.
[36] Kitagawa G. Monte Carlo Filter and Smoother for Non-Gaussian Nonlinear StateSpace Models[J]. Computational and Graphical Statistics,1996,5:1-25.
[37] Doucet A, Godsill S J, Andrieu C. On Sequential Monte Carlo Sampling Methods forBayesian Filtering[J]. Statistics and Computing,2000,10(3):197-208.
[38] Isard M, Blake A. Contour Tracking by Stochastic Propagation of ConditionalDensity[A]. Proceedings of European Conference on Computer Vision[C]. Berlin:Springer-Verlag,1996:343-356.
[39] Perez P, Hue C, Vermaak J, et al. Color-based Probabilistic Tracking[A]. Proceedingsof European Conference on Computer Vision[C]. Berlin: Springer-Verlag,2002:661-675.
[40] Nummiaro K, Koller-Meier E, Gool L V. An Adaptive Color-based Particle Filter[J].Image and Vision Computing,2003,21(1):100-110.
[41] Wu Y, Huang T S. Robust Visual Tracking by Integrating Multiple Cues Based onCo-inference Learning[J]. International Journal of Computer Vision,2004,58(1):55-71.
[42] Jacquot A, Sturm P, Ruch O. Adaptive Tracking of Non-rigid Objects Based on ColorHistograms and Automatic Parameter Selection[J]. IEEE Workshop on Motion andVideo Computing, New York: IEEE Press,2005:103-109.
[43] Stauffer C, Grimson W. Learning Patterns of Activity Using Real-time Tracking[J].IEEE Trans. Pattern Anal. Machine Intelligence,2000,22(8):747-757.
[44] McKenna S J, Gong S, Raja Y. Modelling Facial Color and Identity with GaussianMixtures[J]. Pattern Recognition.1998,31(12):1883-1892.
[45] Isard M, Blake A. Icondensation: Unifying Low-level and High-level Tracking in aStochastic Framework[A]. Proceedings of European Conference on ComputerVision[C]. Berlin: Springer-Verlag,1998:893-908.
[46] Spengler M, Schiele B. Towards Robust Multi-cue Integration for Visual Tracking[J].Machine Vision and Applications,2003,14(1):50-58.
[47] Okuma K, Taleghani A, Nando de Freitas, et al. A Boosted Particle Filter: MultitargetDetection and Tracking[A]. Proceedings of European Conference on ComputerVision[C]. Berlin: Springer-Verlag,2004:28-39.
[48] Yang C, Durraiswami R, Davis L S. Fast Multiple Object Tracking Via a HierarchicalParticle Filter[A]. International Conference on Computer Vision[C], New York: IEEEPress,2005:212-219
[49] Viola P, Jones M J. Rapid M J. Rapid Object Detection Using a Boosted Cascade ofSimple Features[A]. IEEE Conference on Computer Vision and Pattern Recognition[C],New York: IEEE Press,2001:511-518.
[50] Jones M, Viola P. Fast Multi-view Face Detection[A]. IEEE Conference on ComputerVision and Pattern Recognition[C]. New York: IEEE Press,2003:587-594.
[51] Ning H Z, Tan T N, Wang L, et al. People Tracking Based on Motion Model andMotion Constraints with Automatic Initialization[J]. Pattern Recognition,2004,37(7):1423-1440.
[52] Wang J Y, Chen X L, Gao W. Online Selecting Discriminative Tracking Features usingParticle Filter[A]. IEEE Conference on Computer Vision and Pattern Recognition[C].New York: IEEE Press,2005:1037-1042.
[53] Wang Q, Ai H Z, Xu G Y. Learning Based Tracking of Ccomplex Non-rigid Motion[J].Journal of Computer Science and Technlogy,2004,19(4):489-500.
[54] Li P H, Zhang T W, Pece A E C. Visual Contour Tracking Based on Particle Filters[J].Image and Vision Computing,2003,21(1):111-123.
[55] K. Fukunaga and L.D. Hostetler. The Estimation of the Gradient of a Density Function,with Applications in Pattern Recognition[J]. IEEE Trans. Information Theory.1975(21)：32-40.
[56] Y. Cheng, Mean Shift, Mode Seeking, and Clustering[J]. IEEE Trans. Pattern Analysisand Machine Intelligence.1995,17(8):790-799.
[57] Comaniciu D, Ramesh V, Meer P. Real-Time Tracking of Non-rigid Objects UsingMean Shift[A]. Proceedings of the IEEE Conference on Computer Vision and PatternRecognition,2000:142-149.
[58] Comaniciu D, Ramesh V, Meer P. Kernel-Based Object Tracking[J]. IEEE Trans. OnPattern Analysis and Machine Intelligence,2003,25(5):564-575.
[59] Lindeberg T. Feature Detection with Automatic Scale Selection[J]. InternationalJournal of Computer Vision,1998,30(2):194-203.
[60] Collins R T. Mean Shift Blob Tracking through Scale Space[A]. IEEE Conference onComputer Vision and Pattern Recognition[C], New York: IEEE Press,2003:234-240.
[61]李乡儒，吴福朝，胡占义.均值漂移算法的收敛性[J].软件学报，2005,16(3):365-374.
[62] Li X R, Hu Z Y, Wu F C. A Note on the Convergence of the Mean Shift[J]. PatternRecognition,2007,40(6):1756-1762.
[63] Zhao Q, Tao H. Object Tracking Using Color Correlogram[A]. Proceedings of IEEEWorkshop on Visual Survellance and Performance Evaluation of Tracking andSurveillance[C], New York: IEEE Press,2005:263-270.
[64] Birchfield S T, Rangarajan S. Spatiograms Versus Histograms for Region-basedTracking[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C], New York: IEEE Press,2005:1158-1163.
[65]李培华.一种改进的MeanShift跟踪算法[J].自动化学报，2007，33(4):347-354.
[66]杜凯，巨永锋，靳引利，李刚.自适应空间颜色直方图的Meanshift跟踪算法[J].武汉理工大学学报，2012,34(6):140-144.
[67] Liu T L, Chen H T. Real-time Tracking Using Trust-region Methods[J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2004,26(3):397-402.
[68] Yang C J, Duraiswami R, de Menthon D, et al. Mean-shift Analysis UsingQuasi-newton Methods[A]. Proceedings of IEEE Conference on Image Processing[C].New York: IEEE Press,2003:447-450.
[69] Fashing M, Tomasi C. Mean Shift is a Bound Optimization[J]. IEEE Transactions onPattern Analysis and Machine Intelligence,2005,27(3):471-474.
[70] Bajramovic F, Grabl C, Denzler J. Efficient Combination of Histograms for Real-timeTracking Using Mean-shift and Trust-region Optimization[A]. Proceedings of27thDAGM Symposium on Pattern Recognition[C], Berlin: Springer-Verlag,2005:254-261.
[71] Zhang H, Huang Z, Huang W, et al. Kernel-based Method for Tracking Objects withRotation and Translation[A]. Proceedings of International Conference on PatternRecognition[C], New York: IEEE Press,2004:728-731.
[72] Zhang H H, Huang W M, Huang Z Y, et al. Affine Object Tracking with Kernel-basedSpatial Color Representation[A]. IEEE Conference on Computer Vision and PatternRecognition[C], New York: IEEE Press,2005:293-300.
[73] Yilmaz A. Object Tracking by Asymmetric Kernel Mean Shift with Automatic Scaleand Orientation Selection[A]. Proceeding of IEEE Conference on Computer Vision andPattern Recognition[C], New York: IEEE Press,2007:1-6.
[74] Elgammal A, Duraiswami R, Davis L S. Probabilistic Tracking in Joint Feature-spatialSpaces[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C], New York: IEEE Press,2003:781-788.
[75] Elgammal A, Duraiswami R, Davis L S. Efficient Non-parametric Adaptive ColorModeling Using Fast Gauss Transform[A]. Proceedings of IEEE Conference onComputer Vision and Pattern Recognition[C]. New York: IEEE Press,2001:781-788.
[76] Yang C J, Duraiswami R, Davis L. Efficient Mean-shift Tracking Via a New SimilarityMeasure[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C], New York: IEEE Press,2005:176-183.
[77] Hager G, Dewan M, Stewart C. Multiple Kernel Tracking with SSD[A]. Proceedingsof IEEE Conference on Computer Vision and Pattern Recognition[C]. New York: IEEEPress,2004:790-797.
[78] Leichter I, Lindenbaum M, Rivlin E. Visual Tracking by Affine Kernel Fitting UsingColor and Object Boundary[A]. Proceedings of International Conference on ComputerVision[C]. New York: IEEE Press,2007:1-6.
[79] Zivkovic Z, Krose B. An EM-like Algorithm for Color-Histogram-Based ObjectTracking[A]. Proceedings of IEEE Conference on Computer Vision, and PatternRecognition[C], New York: IEEE Press,2004:798-803.
[80] Shen C H, Brooks M J, Hegnel A. Fast Global Kernel Density Mode Seeking:Applications to Localization and Tracking[J]. IEEE Transaction on Image Processing,2007,16(5):1457-1469.
[81] Zhao Q, Brennan S, Tao H, et al. Differential EMD tracking[A]. Proceedings ofInternational Conference on Computer Vision[C]. New York: IEEE Press,2007:1-8.
[82] Han B, Comaniciu D, Zhu Y, et al. Sequential Kernel Density Approximation and itsApplication to Real-time Visual Tracking[J]. IEEE Transaction on Pattern Analysis andMachine Intelligence,2008,30(7):1186-1197.
[83] Park M, Liu Y, Collins R. Efficient Mean Shift Belief Propagation for VisionTracking[A]. Proceedings of IEEE Conference Computer Vision and PatternRecognition[C]. New York: IEEE Press,2008:1-8.
[84] Fan Z, Yang M, Wu Y. Multiple Collaborative Kernel Tracking[J]. IEEE Transactionon Pattern Analysis and Machine Intelligence,2007,29(7):1268-1273.
[85] Kass M, Witkings A, Terzopoulos. Snakes: Active Contour Models[J]. InternationalJournal of Computer Vision,1988,1(4):321-331.
[86] Aubert G, Kornprobst P. Mathematical Problems in Image Processing: PartialDifferential Equations and the Calculus of Variations[M]. Berlin: Springer-Verlag,2006.
[87] Caselles V, Kimmel R, Sapiro G. Geodesic active contours[J]. International Journal ofComputer Vision,1997,22(3):61-79.
[88] Paragios N, Deriche R. Geodesic Active Regions and Level Set Methods for MotionEstimation and Tracking[J]. Computer Vision and Image Understanding,2005,97(3):259-282.
[89] Paragios N, Deriche R. Geodesic Active Contours and Level Set for the Detection andTracking of Moving Objects[J]. IEEE Trans. Pattern Anal. Machine Intell.,2000,22(3):266-280.
[90] Boltz S, Herbulot A, Debreuve E, et al. Motion and Appearance Nonparametric JointEntropy for Video Segmentation[J]. International Journal of Computer Vision,2008,80(2):242-259.
[91] Jehan-Besson S, Barlaud M, Aubert G, et al. Shape Gradients for HistogramSegmentation Using Active Contours[A]. International Conference on ComputerVision[C]. New York: IEEE Press,2003:408-415.
[92] Freedman D, Zhang T. Active Contours for Tracking Distributions[J]. IEEE Trans.Image Processing,2004,13(4):518-526.
[93] Cremers D. Nonlinear Dynamical Shape Priors for Level Set Segmentation[A]. IEEEConference on Computer Vision and Pattern Recognition[C], New York: IEEE Press,2007:1-7.
[94] Sundaramoorthi G, Yezzi A, Mennucci A, et al. New Possibilities with Sobolev ActiveContours[J]. International Journal of Computer Vision, DOI:10.1007/s11263-008-0133-9,2009.
[95] Chan T F, Vese L A. Active Contours without Edges[J]. IEEE Trans. Image Processing,2001,10(2):266-277.
[96] Rathi Y, Vaswani N, Tannenbaum A, et al. Tracking Deforming Objects Using ParticleFiltering for Geometric Active Contours[J]. IEEE Trans. Pattern Anal. MachineIntelligence,2007,29(8):1470-1475.
[97] Cremers D, Rousson M, Deriche R. A Review of Statistical Approaches to Level SetSegmentation: Integrating Color, Texture, Motion and Shape[A]. International Journalof Computer Vision[C],2007,72(2):195-215.
[98] Aubert G, Barlaud M, Faugeras O, et al. Image Segmentation Using Active Contours:Calculus of Variations or Shape Gradients?[J]. SIAM Journal of Applied Mathematics,2003,63(6):2128-2154.
[99] Jehan-Besson S, Barlaud M, Aubert G. DREAM2S: Deformable Regions Driven by anEulerian Accurate Minimization Method for Image and Video Segmentation[J].International Journal of Computer Vision,2003,53(1):45-70.
[100] Osher S, Sethian J A. Fronts Propagating with Curvature-Dependent Speed:Algorithms Based on Hamilton-Jacobi Formulations[J]. Journal of ComputationalPhysics,1988,79:12-49.
[101] Sethian J A. Level Set Methods and Fast Marching Methods[M]. Cambridge:Cambridge Universty Press,1999.
[102] Goldenberg R, Kimmel R, Rivlin E, et al. Fast Geodesic Active Contours[J]. IEEETrans. on Image Processing,2001,10(10):1467-1475.
[103] Papandreou G, Maragos P. Multigrid Geometric Aactive Contour Models[J]. IEEETrans. on Image Processing,2007,16(1):229-240.
[104]马波，张田文.基于曲线进化的目标检测算法[J].信号处理，2003,3:237-241.
[105]于慧敏，徐艺，刘继忠，等.基于水平集的多运动目标时空分割与跟踪[J].中国图像图形学报，2007,7:1218-1223.
[106]周则明，王元全，王平安，等.基于水平集的3D左心室表面重建[J].计算机研究与发展,2005,42(7):1173-1178.
[107]陈卫刚，戚飞虎.一种改进的窄带算法在彩色序列图像人脸跟踪中的应用[J].电子与信息学报，2005,4:540-543.
[108]黄福珍，苏剑波.基于Level Set方法的人脸轮廓提取与跟踪[J].计算机学报，2003,26(4):491-496.
[109]王元全，贾云得.一种新的心脏核磁共振图像分割方法[J].计算机学报，2007,30(1):129-136.
[110]查宇飞，张育，毕笃彦.基于区域活动轮廓运动目标跟踪方法研究[J].中国图像图形学报，2006,12:1844-1848.
[111]罗嘉，韦志辉.基于几何活动轮廓模型的目标跟踪与快速运动估计[J].中国图像图形学报，2009,14(7):1361-1368.
[112]周雪，胡卫明.融合颜色和增量形状先验的目标轮廓跟踪[J].自动化学报,2009,35(11):1394-1402.
[113] Li P H, Xiao L J. Histogram-Based Partial Differential Equation for ObjectTracking[A]. The7th International Conference on Advances in Pattern Recognition,ICAPR09[C]. New York: IEEE Press,2009:286-289.
[114] Jepson A. D., Fleet D, J., El-Maraghi T. F. Robust Online Appearance Models forVisual Tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(10):415-522.
[115] Ross D., Lim J., Yang M.-H. Adaptive Probabilistic Visual Tracking with IncrementalSubpace Update[A]. Proceedings of the eighth European conference on computervision[C]. Berlin: Springer,2004.
[116] Wu Y., Huang T. S. Robust Visual Tracking by Integrating Multiple Cues Based onCo-Inference Learning[J]. International Journal of Computer Vision.2004,58(1):55-71.
[117] Cannons K., Wildes R. Spatiotemporal Oriented Energy Features for VisualTracking[A]. In: Asian Conference on Computer Vision[C].2007. pp532-543.
[118] Senior A., Hampapur A., Tian Y.-L., et al. Appearance Models for OcclusionHandling[J]. Image and Vision Computing,2006,24(11):1233-1243.
[119] Yilmaz A., Xin L., Shah M. Contour-based Object Tracking with Occlusion Handlingin Video Acquired Using Mobile Cameras[J]. IEEE Transcations on Pattern Analysisand Machine Intelligence,2004,26(11):1531-1536.
[120] Tao Y., Quan P., Jing L., et al. Real-time Multiple Objects Tracking with OcclusionHandling in Dynamic Scenes[A]. Proceedings of IEEE Conference on Computer Visionand Pattern Recognition[C]. San Diego, CA, USA: IEEE,2005:970-975.
[121] Onoguchi K. Shadow Elimination Method for Moving Object Detection[A].Proceedings of IEEE International Conference on Pattern Recognition[C]. New York:IEEE Press,1998,1:583-587.
[122] Comaniciu, D. and P. Meer. Mean Shift Analysis and Application[A]. Proceedings ofthe7th IEEE International Conference on Computer Vision[C]. Kerkyra, Greece,1999,pp1197-1203.
[123] Comaniciu, D. and P. Meer. Mean Shift: A Robust Application toward Feature SpaceAnalysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603-619.
[124] Comaniciu, D. and P. Meer. Robust Analysis of Feature Spaces: Color ImageSegmentation[A]. Proceedings of IEEE Conference Computer Vision and PatternRecognition[C]. San. Juan, Puerto Rico1997:750–755.
[125] Comaniciu, D., V. Ramesh, and P. Meer. The Variable Bandwidth Mean Shift andData-driven Scale Selection[A]. Proceedings of the8th International Conference onComputer Vision[C]. Vancouver, Canada,2001:438-445.
[126] Yang, C,. R. Duraiswami, and L. Davis. Similarity Meansure for NonparametricKernel Density Based Object Tracking[A]. Eighteenth Annual Conference on NeuralInformation Processing Systems[C]. Victoria, British Columbia, Canada: NIPS,2004.
[127] Collins, R.T. Mean-Shift Blob Tracking through Scale Space[A]. in IEEE Conferenceon Computer Vision and Pattern Recognition[C].2003, Madison, Wisconsin,2003:234–240.
[128] Nummiaro, K., E. Koller-Meier, and L. Van Goo. Color Features for TrackingNon-Rigid Objects[J]. Chinese Journal of Automation,2003,29(3):345-355.
[129] McFarlane, N.J.B. and C.P. Schofield. Segmentation and Tracking of Piglets inImages [J]. Machine Vision and Applications,1995,8:187-193.
[130] Kailath, T., The Divergence and Bhattacharyya Distance Measures in SignalSelection [J]. IEEE Trans. Comm. Technology.,1967.15:52-60.
[131] Swain, M.J. and D.H. Ballard, Color Indexing[J]. International Journal of ComputerVision,1991.7(1):11-32.
[132] Maggio, E. and A. Cavallaro. Hybrid Particle Filter and Mean Shift Tracker withAdaptive Transition Model[A]. In Acoustics, Speech, and Signal Processing,2005.Proceedings[C].2005.
[133]袁亚湘，孙文瑜.最优化理论与方法[M].北京：科学出版社，2001.
[134] David K, Ward C. Numberical Analysis: Mathematics of Scientific Computing[M].Washington: American Mathematical Society,2002.
[135] Nocedal J, Wright S J. Numerical Optimization[M]. New York：Springer-Verlag,1999.
[136] D. Xu, Y. Wang, and J. An. Applying a New Spatial Color Histogram in Mean-ShiftBased Tracking Algorithm[A]. Image and Vision Computing[C]. New Zealand,2005.
[137] E. Maggio, A. Cavallaro. Multi-Part Target Representation for Color Tracking[A].Proc. IEEE International Conference on Image Processing[C],2005:729-732.
[138]袁霄,王丽萍.基于MeanShift算法的运动人体跟踪[J].计算机工程与科学,2008,30(4):46-49.
[139]张恒,李由,李立春,于起峰.基于显著性加权的Mean Shift跟踪方法[J].光学技术,2008,34(3):404-407.
[140]胡铟,杨静宇.基于分块颜色直方图的MeanShift跟踪算法[J].系统仿真学报,2009,21(10):2936-2955.
[141]王东峰，张丽飞，刘小军，邹谋炎.基于广义特征点匹配的全自动图像配准[J].电子与信息学报，2005，27(7):1013-1016.
[142]王永明，王贵锦.图像局部不变性特征与描述[M].北京：国防工业出版社，2010.
[143] Lindeberg T. Detecting Salient Blob-Like Image Structures and Their Scales with aScale-Space Primal Sketch: A Method for Focus-of-Attention[J]. International Journalof Computer Vision,1993,11(3).
[144] Lindeberg T. Feature Detection with Automatic Scale Selection[J]. InternationalJournal of Computer Vision,1998,30(2).
[145] Lowe D G. Distinctive Image Features from Scale-invariant Keypoints[J].International Journal of Computer Vision,2004,60(2).
[146] Bay H, Tuytelaars T, Van Gool L. SURF: Speeded up Robust Features[C]. In ECCV,2006.
[147] Bay H, Tuytelaars T, Van Gool L. Speeded-Up Robust Features(SURF)[J]. ComputerVision and Image Understanding,2008.
[148] Bay H. From Wide-baseline Point and Line Correspondences to3D[D]. Ph.D.Thesis,ETH Zurich,2006.
[149] Bay H, Fasel B, Van Gool L. Interactive Museum Guide: Fast and RobustRecognition of Museum Objects[A]. Proceedings of the First International Workshopon Mobile Vision,2006.
[150] Viola P, Jones M. Rapid Object Detection Using a Boosted Cascade of simplefeatures[C]. In IEEE Conference on Computer Vision and Pattern Recoghition,2001,1.
[151] Simard P, Bottou L, Haffner P, et.al. Boxlets: a Fast Convolution Algorithm forSignal Processing and Neural Networks[J]. Advances in Neural Information ProcessingSystems,1999,11.
[152] K. Eugene. Feature Descriptor Comparison Report[EB/OL]. http://computer-vision-talks.com/2011/08/feature-descriptor-comparison-report/,2011.
[153] M. Calonder, V. Lepetit, C. Strecha, P. Fua. BRIEF: Binary Robust IndependentElementary Features[C]. In ECCV2010.
[154] E. Rublee, V. Rabaud, K. Konolige, G. Bradski. ORB: an Efficient Alternative toSIFT or SURF[C]. In ICCV2011.
[155] Zhou H Y, Yuan Y, and Shi C. Object Tracking Using SIFT Features and MeanShift[J]. Computer Vision and Image Understanding,2009,345-352.
[156]李培华,肖莉娟.基于Mean Shift的相似性变换和仿射变换目标跟踪算法[J].中国图象图形学报,2011,16(2):258-266.
[157]张贤达.矩阵分析与应用[M].北京：清华大学出版社,2004.
[158] Heikkila M, Pietikainen M. A Texture-based Method for Modelling the Backgroundand Detecting Moving Objects[J]. IEEE Transaction on Pattern Analysis and MachineIntelligence,2006,28(4):657-662.