视频序列图像中运动目标检测与跟踪算法研究
|
摘要
随着计算机技术的不断进步以及计算机运算能力的持续提升，计算机视觉作为模拟人类视觉功能的复杂课题受到了越来越多的关注。视频序列图像中的运动目标检测与跟踪技术是计算机视觉的一个十分重要的研究方向，其目的是在序列图像中根据信息在时间和空间上的相关性确定目标在每一帧的位置、形态和速度等属性，它对后续基于视觉的应用任务具有重要的意义。由于受到了目标自身变化、环境光影改变、成像设备噪声等多方面干扰的影响，使得设计一个兼具鲁棒性、准确性和快速性的检测与跟踪算法依然是一项极具挑战的开放性课题。本文在调研已有研究成果的基础上，探索了背景建模、聚类定位、多特征融合、轮廓演化模型以及滤波估计框架等相关技术领域，创新地提出了针对一些具体问题的解决方法。主要研究工作和结论如下：
首先，为提高目标检测系统中高斯背景模型的更新速度，提出了场景运动复杂度的概念和计算方法，并在此基础上提出了一种组合高斯模型的背景建模方法。根据像素的时空采样模型分析场景运动的复杂性并计算出场景的熵值图，按照最大熵阈值将熵值图分割为稳定区域和动态区域，然后在不同的区域采用不同的高斯模型和更新算法。与固定模态个数的高斯模型相比，由于它是建立在对背景区域做出合理分析和解释之后一种组合建模方法，能够避免高斯模态的浪费，又提高了背景参数的更新速度和前景目标的检出速度。
其次，提出了一种基于聚类分析的目标定位算法。由于检出的前景常具有野点数据多、区域连通性差等问题，在利用区域生长法检出前景目标属性（大小与位置）时，常会出现错误的检出结果。为此提出了基于近邻分析的前景目标定位算法：首先，经过对背景减法所得到的前景进行适当的下采样和滤波操作后，将目标定位问题转化为像素聚类问题；然后，基于对这些像素间的距离和上下文关系的分析提出了互邻特征矩阵，定义了聚类准则函数；最后，在准则函数最小化标准下设计了完整的聚类算法。通过上述近邻分析过程将前景像素归为几个特定的聚类，从而实现对前景目标的准确定位。
然后，提出两种基于区域统计特征的跟踪算法。（1）在无迹卡尔曼滤波框架下，提出了一种基于颜色和边缘统计特征融合的目标跟踪算法。结合了区域的颜色和边缘特征的表观模型比单一区域特征的表观模型更加全面准确地表征了目标，提高了跟踪算法的抗干扰能力和准确性。与此同时，UKF高效的预测更新机制减少了均值漂移算法迭代的次数，提高了目标的搜索效率。（2）在粒子滤波框架下，提出了一种基于颜色与光流特征的目标跟踪算法。为了能够确定目标坐标、速度、大小以及旋转等多方面的信息，并且为了克服单一颜色特征表征造成的可分性差的问题，首次在层次粒子框架下提出了的基于颜色和运动特征的跟踪算法。实验表明，由于采用了层次结构设计和颜色与光流两种区域特征，算法能够自适应地调整跟踪窗口的位置、大小和方向，正确地估计目标状态，得到了比较准确的跟踪结果。在有遮挡时，算法能正确地预测目标位置并在目标重新出现后能够及时捕捉目标继续跟踪，表现了较好的鲁棒性。
最后，在UKF框架下，提出了一种基于边缘特征的目标跟踪算法。在经典的几何式主动轮廓分割算法中，为了得到准确的分割结果需要冗长的迭代过程，而且有时并不能如愿。为了提高分割效率和准确性，在UKF框架下设计了全新边缘检测与跟踪算法：首先，采用矢量图像计算图像的梯度值，并设计了能够自适应调整阈值的边缘指示函数；然后，提出了改进后的变分水平集演化模型；最后，在UKF框架下设计了针对运动目标的边缘检测与跟踪算法。实验证明，算法不但显著地提高了轮廓演化模型的灵活性和收敛速度，而且对于阴影、遮挡、目标形变和背景干扰等表现出了较好的鲁棒性。
总之，全文针对本课题的有关问题进行了比较系统和深入的研究，相关实验表明，所提出算法在鲁棒性和准确性方面都有较大提高。
With the development of computer technology and improvement of computing power, acomplicated research topic, computer vision, attracts more and more attention, which aims tosimulate human visual capabilities. As a very important research interest of computer vision,the detection and tracking of moving objects in sequence images is to recognize the objects’information such as position, shape, velocity, etc. based on their temporal and spatialcorrelation, which has great significance to further vision-based applications. The detectionand tracking technology is facing many challenges such as object defomation, changefulcircumstance, unstable imaging devices, etc., which makes it still a challenging open researchissue to devise a robust, accurate and rapid detection and tracking algorithm. The dissertationexplored the concerned technologies by investigating the existing work such as backgroundmodeling, locating based on clustering, multi-feature fusion, contour evolution, filteringestimation framework, etc. and proposed innovative solutions to some related problems. Itsmain work and conclutions are as follows:
Firstly, in order to improve the updating speed of Gaussian models for background, theconcept and computation method of the scene moving complexity were devised, according towhich a combinational Gaussian model for background modeling was proposed. In thismethod, according to the spatio-temporal sampling model of pixels, the scene movingcomplexity was analyzed and the entropy image of the scene was calculated. And this imagewas segmented into the stable region and the dynamic region by means of the maximumentropy threshold. In the two different regions, two different Gaussian models andcorresponding updating algorithms were respectively adopted. Since the modeling method isbased on reasonable analysis and classification for the surveillance scene, the proposed modelcan avoid the waste of Gaussians and be provided with higher updating and detecting speedcomparing to fixed number of Gaussians.
Secondly, a object locating algorithm was proposed based on clustering analysis.Because the foreground resulted from background subtract often has many undesirable features such as abundance of outliers, spoiled connectivity, etc., the furter detection byreagion growing almost can’t locate the outline of the object correctlly. In order to solve theseproblems, a locating algorithm of moving objects was proposed based on neighboringanalysis. First, the foreground resulted from background subtract was downsampled andfiltered. And the locating of moving objects was converted to pixels clustering. Second, theneighboring feature matrix and the criterion function were proposed based on analysis of thedistance and context of the pixels. Finally, according to the minimization of the criterionfunction the clustering algorithm was devised. And the pixels were clustered into a certainnumber of clusters corresponding to objects. Thus, the foreground objects were locatedcorrectly.
Then, the two tracking algorithms were presented based on in the framework of UKF aobject tracking algorithm was presented based on the region statistical characteristics.(1) Inthe framework of UKF a improved tracking algorithm based on the fusion of color and edgefeatures was proposed. The multi-feature fusion appearance model describes the object morecomprehensively than the single feature one, which has enhanced the accuracy and adaptivityof the tracking algorithm. At the same time, because of the effective “predict-update”mechanism in UKF, the iterations of mean-shift has greatly decreased and the searchefficiency for the tracked object is improved.(2) In the framework of Particle filtering, aobject tracking algorithm were presented based on region statistical characteristics of colorand optical-flow. In order to recognize various information of the object such as coordinates,speed, size and rotation, and to overcome the poor discrimination of color features, a trackingalgorithm in a hierarchical Particle filtering framework was proposed for the first time. Itadopted color and optical-flow features. The tests indicated it could adaptively adjust thetracking window’s position, size and rotation, estimat the target’s state correctly and track theobject accurately because of adopting hierarchical filtering stracture and two-feature fusionalgorithm. And it predicted the object’s position correctly under occlusion and catched itshortly when the occlusion disappeared, which showed it had a robust performance.
Finally, in the UKF framework, a object tracking algorithm was proposed based on edge feature. During executing the algorithm of classical geometric active contour segmentation, toobtain accurate segmentation results always involves lengthy iterative process, which doesn'teven work. To improve the segmentation efficiency and accuracy, a novel detection andtracking algorithm was presented. First, the gradient image was calculated based on the vectorimage and an adaptive edge indicator was proposed. Second, the revised evolution modelusing variational level set method was put forward. And then the detection and tracking of theobject's edge is presented in the framework of UKF. The experiments demonstrate not onlythat it has significantly increased the convergence rate and flexibility of the active contourevolution but also that it is robust to some interference such as shadow, occlusion,deformation of object and background interference.
In summary, the dissertation has been presented in a comprehensive way to discuss theserelevant issues. And extensive experiments shows the proposed algorithms have been greatlyimproved in terms of roubustness and accuracy.
首先，为提高目标检测系统中高斯背景模型的更新速度，提出了场景运动复杂度的概念和计算方法，并在此基础上提出了一种组合高斯模型的背景建模方法。根据像素的时空采样模型分析场景运动的复杂性并计算出场景的熵值图，按照最大熵阈值将熵值图分割为稳定区域和动态区域，然后在不同的区域采用不同的高斯模型和更新算法。与固定模态个数的高斯模型相比，由于它是建立在对背景区域做出合理分析和解释之后一种组合建模方法，能够避免高斯模态的浪费，又提高了背景参数的更新速度和前景目标的检出速度。
其次，提出了一种基于聚类分析的目标定位算法。由于检出的前景常具有野点数据多、区域连通性差等问题，在利用区域生长法检出前景目标属性（大小与位置）时，常会出现错误的检出结果。为此提出了基于近邻分析的前景目标定位算法：首先，经过对背景减法所得到的前景进行适当的下采样和滤波操作后，将目标定位问题转化为像素聚类问题；然后，基于对这些像素间的距离和上下文关系的分析提出了互邻特征矩阵，定义了聚类准则函数；最后，在准则函数最小化标准下设计了完整的聚类算法。通过上述近邻分析过程将前景像素归为几个特定的聚类，从而实现对前景目标的准确定位。
然后，提出两种基于区域统计特征的跟踪算法。（1）在无迹卡尔曼滤波框架下，提出了一种基于颜色和边缘统计特征融合的目标跟踪算法。结合了区域的颜色和边缘特征的表观模型比单一区域特征的表观模型更加全面准确地表征了目标，提高了跟踪算法的抗干扰能力和准确性。与此同时，UKF高效的预测更新机制减少了均值漂移算法迭代的次数，提高了目标的搜索效率。（2）在粒子滤波框架下，提出了一种基于颜色与光流特征的目标跟踪算法。为了能够确定目标坐标、速度、大小以及旋转等多方面的信息，并且为了克服单一颜色特征表征造成的可分性差的问题，首次在层次粒子框架下提出了的基于颜色和运动特征的跟踪算法。实验表明，由于采用了层次结构设计和颜色与光流两种区域特征，算法能够自适应地调整跟踪窗口的位置、大小和方向，正确地估计目标状态，得到了比较准确的跟踪结果。在有遮挡时，算法能正确地预测目标位置并在目标重新出现后能够及时捕捉目标继续跟踪，表现了较好的鲁棒性。
最后，在UKF框架下，提出了一种基于边缘特征的目标跟踪算法。在经典的几何式主动轮廓分割算法中，为了得到准确的分割结果需要冗长的迭代过程，而且有时并不能如愿。为了提高分割效率和准确性，在UKF框架下设计了全新边缘检测与跟踪算法：首先，采用矢量图像计算图像的梯度值，并设计了能够自适应调整阈值的边缘指示函数；然后，提出了改进后的变分水平集演化模型；最后，在UKF框架下设计了针对运动目标的边缘检测与跟踪算法。实验证明，算法不但显著地提高了轮廓演化模型的灵活性和收敛速度，而且对于阴影、遮挡、目标形变和背景干扰等表现出了较好的鲁棒性。
总之，全文针对本课题的有关问题进行了比较系统和深入的研究，相关实验表明，所提出算法在鲁棒性和准确性方面都有较大提高。
With the development of computer technology and improvement of computing power, acomplicated research topic, computer vision, attracts more and more attention, which aims tosimulate human visual capabilities. As a very important research interest of computer vision,the detection and tracking of moving objects in sequence images is to recognize the objects’information such as position, shape, velocity, etc. based on their temporal and spatialcorrelation, which has great significance to further vision-based applications. The detectionand tracking technology is facing many challenges such as object defomation, changefulcircumstance, unstable imaging devices, etc., which makes it still a challenging open researchissue to devise a robust, accurate and rapid detection and tracking algorithm. The dissertationexplored the concerned technologies by investigating the existing work such as backgroundmodeling, locating based on clustering, multi-feature fusion, contour evolution, filteringestimation framework, etc. and proposed innovative solutions to some related problems. Itsmain work and conclutions are as follows:
Firstly, in order to improve the updating speed of Gaussian models for background, theconcept and computation method of the scene moving complexity were devised, according towhich a combinational Gaussian model for background modeling was proposed. In thismethod, according to the spatio-temporal sampling model of pixels, the scene movingcomplexity was analyzed and the entropy image of the scene was calculated. And this imagewas segmented into the stable region and the dynamic region by means of the maximumentropy threshold. In the two different regions, two different Gaussian models andcorresponding updating algorithms were respectively adopted. Since the modeling method isbased on reasonable analysis and classification for the surveillance scene, the proposed modelcan avoid the waste of Gaussians and be provided with higher updating and detecting speedcomparing to fixed number of Gaussians.
Secondly, a object locating algorithm was proposed based on clustering analysis.Because the foreground resulted from background subtract often has many undesirable features such as abundance of outliers, spoiled connectivity, etc., the furter detection byreagion growing almost can’t locate the outline of the object correctlly. In order to solve theseproblems, a locating algorithm of moving objects was proposed based on neighboringanalysis. First, the foreground resulted from background subtract was downsampled andfiltered. And the locating of moving objects was converted to pixels clustering. Second, theneighboring feature matrix and the criterion function were proposed based on analysis of thedistance and context of the pixels. Finally, according to the minimization of the criterionfunction the clustering algorithm was devised. And the pixels were clustered into a certainnumber of clusters corresponding to objects. Thus, the foreground objects were locatedcorrectly.
Then, the two tracking algorithms were presented based on in the framework of UKF aobject tracking algorithm was presented based on the region statistical characteristics.(1) Inthe framework of UKF a improved tracking algorithm based on the fusion of color and edgefeatures was proposed. The multi-feature fusion appearance model describes the object morecomprehensively than the single feature one, which has enhanced the accuracy and adaptivityof the tracking algorithm. At the same time, because of the effective “predict-update”mechanism in UKF, the iterations of mean-shift has greatly decreased and the searchefficiency for the tracked object is improved.(2) In the framework of Particle filtering, aobject tracking algorithm were presented based on region statistical characteristics of colorand optical-flow. In order to recognize various information of the object such as coordinates,speed, size and rotation, and to overcome the poor discrimination of color features, a trackingalgorithm in a hierarchical Particle filtering framework was proposed for the first time. Itadopted color and optical-flow features. The tests indicated it could adaptively adjust thetracking window’s position, size and rotation, estimat the target’s state correctly and track theobject accurately because of adopting hierarchical filtering stracture and two-feature fusionalgorithm. And it predicted the object’s position correctly under occlusion and catched itshortly when the occlusion disappeared, which showed it had a robust performance.
Finally, in the UKF framework, a object tracking algorithm was proposed based on edge feature. During executing the algorithm of classical geometric active contour segmentation, toobtain accurate segmentation results always involves lengthy iterative process, which doesn'teven work. To improve the segmentation efficiency and accuracy, a novel detection andtracking algorithm was presented. First, the gradient image was calculated based on the vectorimage and an adaptive edge indicator was proposed. Second, the revised evolution modelusing variational level set method was put forward. And then the detection and tracking of theobject's edge is presented in the framework of UKF. The experiments demonstrate not onlythat it has significantly increased the convergence rate and flexibility of the active contourevolution but also that it is robust to some interference such as shadow, occlusion,deformation of object and background interference.
In summary, the dissertation has been presented in a comprehensive way to discuss theserelevant issues. And extensive experiments shows the proposed algorithms have been greatlyimproved in terms of roubustness and accuracy.
引文
[1]高文,陈熙霖.计算机视觉、算法与系统原理[M].北京:清华大学出版社,1998:1-20
[2] Collins R T, Lipton A J, Kanade T A System for Video Surveillance and Monitoring:VSAM final report[R]. Robotic Institute Carnegie Mellon University,2000
[3] Remagnino P, Tan T, Baker K. Multi-agent visual surveillance of dynamic scenes[J].Image and Vision Computing,1998,16(8):529-532
[4] Maggioni C, K mmerer B, Gesturecomputer－history, design and applications[A].Cippola R, Pentland A. In Computer Vision for Human-Machine Interaction[M],Cambridge University Press: Cambridge,1998:23-52
[5] Freeman W T, Weissman C. Television control by hand gestures[A]. Proceedings ofInternational Workshop on Automatic Face and Gesture Recognition[C].1995:179-183
[6] Haritaoglu I, Harwood D, Davis L S. W4: real-time surveillance of people and theiractivities[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):809-830
[7] Pavlidis I, Morellas V, Tsiamyrtzis P, et al. Urban surveillance systems: from thelaboratory to the commercial world[J]. Proceedings of the IEEE,2001,89(10):1478-1497
[8] Bogaert M, Chelq N, Cornez P, et al. The PASSWORDS Project (intelligent video imageanalysis system)[A]. Proceedings of International Conference on Image Processing[C].IEEE,1996,3:675-678
[9] Wren C R, Azarbayejani A, Darrell T, et al. Pfinder: Real-time tracking of the humanbody[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19(7):780-785
[10] Coifman B, Beymer D, McLauchlan P, et al. A real-time computer vision system forvehicle tracking and traffic surveillance[J]. Transportation Research Part C: EmergingTechnologies,1998,6(4):271-288
[11] Siebel N T, Maybank S. The ADVISOR visual surveillance system[A]. Proceedings ofthe ECCV2004workshop Applications of Computer Vision[C]. Prague, Czech Republic:2004,1:103-111
[12] Reardon S, FBI launches$1billion face recognition project [N]. New Scientist.7September,2012,2880:20
[13] Arif M, Samani H, Yang C-Y, et al. Adaptation of mobile robots to intelligentvehicles[A]. Proceedings of the4th IEEE International Conference on SoftwareEngineering and Service Science [C]. Beijing: IEEE,2013:550-553
[14] Jiang G-Q, Zhao C-J. Apple recognition based on machine vision[A]. Proceedings ofInternational Conference on Machine Learning and Cybernetics[C]. IEEE,2012,3:1148-1151
[15] Jiang G-Q, Zhao C-J, Si Y-S. A machine vision based crop rows detection foragricultural robots[A]. Proceedings of International Conference on Wavelet Analysis andPattern Recognition[C]. IEEE,2010:114-118
[16] Yesin K, Nelson B J, Papanikolopoulos N P, et al. Active video system for a miniaturereconnaissance robot[A]. Proceedings of IEEE International Conference on Robotics andAutomation[C]. IEEE,2000,4:3919-3924
[17] Wu Y, Huang T. Vision-based gesture recognition: A review[J]. Gesture-basedcommunication in human-computer interaction,1999:103-115
[18] Koh E, Won J, Bae C. Vision-based Virtual Touch Screen Interface[A]. Proceedings ofInternational Conference on Consumer Electronics[C]. IEEE,2008:1-2
[19] Pavlovic V I, Sharma R, Huang T S. Visual interpretation of hand gestures forhuman-computer interaction: A review[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,1997,19(7):677-695
[20] Kwok K W, Yam Y, Lo K W. Vision system and projective rectification for a robotdrawing platform[A]. Proceedings of International Conference on Control andAutomation[C]. IEEE,2005,2:691-696
[21] Zhang J, Sui L, Zhuo L, et al. Pornographic image region detection based on visualattention model in compressed domain[J]. IET Image Processing,2013,7(4):384-391
[22] Neuman M R, Baura G D, Meldrum S, et al. Advances in medical devices and medicalelectronics[J]. Proceedings of the IEEE,2012,100(Special Centennial Issue):1537-1550
[23] Kaiser M, Gans N, Dixon W. Vision-based estimation for guidance, navigation, andcontrol of an aerial vehicle[J]. IEEE Transactions on Aerospace and Electronic Systems,2010,46(3):1064-1077
[24] Huang H-B, Huo H, Fang T. Hierarchical Manifold Learning With Applications toSupervised Classification for High-Resolution Remotely Sensed Images[J]. IEEETransactions on Geoscience and Remote Sensing,2014,52(3):1677-1692
[25] Stauffer C, Grimson W. Adaptive background mixture models for real-time tracking[A].Proceedings of IEEE Computer Society Conference on Computer Vision and PatternRecognition[C]. Fort Collins: IEEE,1999:246-252
[26] Power P W, Schoonees J A. Understanding background mixture models for foregroundsegmentation[A]. Proceedings of Image and Vision Computing[C]. New Zealand:University of Auckland,2002,2002:267-271
[27] KaewTraKulPong P, Bowden R, An improved adaptive background mixture model forreal-time tracking with shadow detection[A]. Jones G A, Paragios N, Regazzoni C S. InVideo-Based Surveillance Systems[M], Springer US:2001,2:135-144
[28] Lee K C, Ho J, Yang M H, et al. Visual tracking and recognition using probabilisticappearance manifolds[J]. Computer Vision and Image Understanding,2005,99(3):303-331
[29] Zivkovic Z. Improved adaptive Gaussian mixture model for background subtraction[A].Proceedings of the17th International Conference on Pattern Recognition[C]. IEEE,2004,2:28-31
[30] Li D, Xu L, Goodman E. Online background learning for illumination-robust foregrounddetection[A]. Proceedings of the11th International Conference on Control AutomationRobotics&Vision[C]. Singapore: IEEE,2010:1093-1100
[31] Bouttefroy P L M, Bouzerdoum A, Phung S L, et al. On the analysis of backgroundsubtraction techniques using Gaussian mixture models[A]. Proceedings of IEEEInternational Conference on Acoustics Speech and Signal Processing[C]. IEEE,2010:4042-4045
[32] Huang T, Qiu J, Sakayori T, et al. Motion detection based on background modeling andperformance analysis for outdoor surveillance[A]. Proceedings of InternationalConference on Computer Modeling and Simulation[C]. IEEE,2009:38-42
[33] Liu Y, Bin Z. The improved moving object detection and shadow removing algorithmsfor video surveillance[A]. Proceedings of International Conference on ComputationalIntelligence and Software Engineering[C]. IEEE,2010:1-5
[34] Lin H-H, Chuang J-H, Liu T-L. Regularized background adaptation: a novel learningrate control scheme for Gaussian mixture modeling[J]. IEEE Transactions on ImageProcessing,2011,20(3):822-836
[35] Elgammal A, Duraiswami R, Harwood D, et al. Background and foreground modelingusing nonparametric kernel density estimation for visual surveillance[J]. Proceedings ofthe IEEE,2002,90(7):1151-1163
[36] Kim K, Chalidabhongse T H, Harwood D, et al. Real-time foreground–backgroundsegmentation using codebook model[J]. Real-time imaging,2005,11(3):172-185
[37] Zhang Z H, Chen R Q, Lu H Q, et al. Moving Foreground Detection Based on ModifiedCodebook[A]. Proceedings of the2nd International Congress on Image and SignalProcessing[C]. IEEE,2009:1-5
[38] Prati A, Mikic I, Trivedi M M, et al. Detecting moving shadows: algorithms andevaluation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(7):918-923
[39] Gibson J J. The perception of the visual world[M]. Houghton Mifflin, England: Oxford,1950:37-91
[40] Horn B K P. Robot vision[M]. MIT Press,1986:278-280
[41] Horn B K P, Schunck B G. Determining optical flow[J]. Artificial intelligence,1981,17(1):185-203
[42] Barron J L, Fleet D J, Beauchemin S S. Performance of optical flow techniques[J].International Journal of Computer Vision,1994,12(1):43-77
[43] Opelt A, Pinz A, Zisserman A. A boundary-fragment-model for object detection[A].Leonardis A, Bischof H, Pinz A. Proceedings of the9th European Conference onComputer Vision[C]. Graz, Austria: Springer Berlin Heidelberg,2006,2:575-588
[44] Dorkó G, Schmid C. Selection of scale-invariant parts for object class recognition[A].Proceedings of the9th IEEE International Conference on Computer Vision[C]. IEEE,2013:634-639
[45] Haralick R M, Shanmugam K, Dinstein I H. Textural features for image classification[J].IEEE Transactions on Systems, Man and Cybernetics,1973,(6):610-621
[46]高隽,谢昭.图像理解理论与方法[M].北京:科学出版社,2009:179-186
[47] Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,2003,25(5):564-577
[48] Hager G D, Belhumeur P N. Efficient region tracking with parametric models ofgeometry and illumination[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1998,20(10):1025-1039
[49] Azarbayejani A, Pentland A P. Recursive estimation of motion, structure, and focallength[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995,17(6):562-575
[50] Isard M, Blake A. Condensation: conditional density propagation for visual tracking[J].International Journal of Computer Vision,1998,29(1):5-28
[51] Han B, Davis L. On-line density-based appearance modeling for object tracking[A].Proceedings of the10th IEEE International Conference on Computer Vision[C]. IEEE,2005,2:1492-1499
[52] Wang H, Suter D, Schindler K, et al. Adaptive object tracking based on an effectiveappearance filter[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(9):1661-1667
[53] Pawar M. Mean shift Face Tracking with Dynamic Target Model update using BayesianSkin Classifier[A]. Proceedings of International Conference on ComputationalIntelligence&Computing Research[C]. Coimbatore: IEEE2012:1-5
[54] Fieguth P, Terzopoulos D. Color-based tracking of heads and other mobile objects atvideo frame rates[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C]. IEEE,1997:21-27
[55] Zoidi O, Tefas A, Pitas I. Visual object tracking based on local steering kernels and colorhistograms[J]. IEEE Transactions on Circuits and Systems for Video Technology,2013,23(5):870-882
[56] Bradski G R. Computer vision face tracking for use in a perceptual user interface[J].Intel Technology Journal Q2,1998,2(2):1-15
[57] Comaniciu D, Ramesh V, Meer P. Real-time tracking of non-rigid objects using meanshift[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C]. IEEE,2000,2:142-149
[58] Pérez P, Hue C, Vermaak J, et al., Color-based probabilistic tracking[A]. In the7thEuropean Conference on Computer Vision[M], Springer: Berlin Heidelberg,2002:661-675
[59] Collins R T. Mean-shift blob tracking through scale space[A]. Proceedings of IEEEComputer Society Conference on Computer Vision and Pattern Recognition[C]. IEEE,2003,2:234-240
[60] Adam A, Rivlin E, Shimshoni I. Robust fragments-based tracking using the integralhistogram[A]. Proceedings of IEEE Computer Society Conference on Computer Visionand Pattern Recognition[C]. IEEE,2006,1:798-805
[61] Stern H, Efros B. Adaptive color space switching for face tracking in multi-coloredlighting environments[A]. Proceedings of the5th IEEE International Conference onAutomatic Face and Gesture Recognition[C]. IEEE,2002:249-254
[62] Collins R T, Liu Y, Leordeanu M. Online selection of discriminative tracking features[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1631-1643
[63] Dalal N, Triggs B. Histograms of oriented gradients for human detection[A].Proceedings of Computer Society Conference on Computer Vision and PatternRecognition[C]. IEEE,2005,1:886-893
[64] Zhu W, Levinson S E. Edge orientation-based multi-view object recognition[A].Proceedings of the15th International Conference on Pattern Recognition[C]. Barcelona:IEEE,2000,2000:1936-1939
[65] Yao B, Liu Z, Nie X, et al. Animated Pose Templates for Modelling and DetectingHuman Actions[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2014,36(3):436-452
[66] Jia W, Hu R-X, Lei Y-K, et al. Histogram of Oriented Lines for PalmprintRecognition[J]. IEEE Transactions on Systems, Man, and Cybernetics,2014,44(3):385-395
[67] Agarwal S, Verma A, Singh P. Content Based Image Retrieval using Discrete WaveletTransform and Edge Histogram Descriptor[A]. Proceedings of International Conferenceon Information Systems and Computer Networks[C]. IEEE,2013:19-23
[68] Lucas B D, Kanade T. An iterative image registration technique with an application tostereo vision[A]. Proceedings of the7th International Joint Conference on ArtificialIntelligence[C].1981:674-679
[69] Frey B J. Filling in scenes by propagating probabilities through layers and intoappearance models[A]. Proceedings of IEEE Conference on Computer Vision andPattern Recognition[C]. IEEE,2000,1:185-192
[70] Olson C F. Maximum-likelihood template matching[A]. Proceedings of IEEEConference on Computer Vision and Pattern Recognition[C]. Hilton Head Island, SC:IEEE,2000,2:52-57
[71] McInerney T, Terzopoulos D. Deformable models in medical image analysis[A].Proceedings of the Workshop on Mathematical Methods in Biomedical ImageAnalysis[C]. IEEE,1996:171-180
[72] Fisker R. Making deformable template models operational[D]. PhD thesis, Copenhagen:Informatics and Mathematical Modelling, Technical University of Denmark,2000
[73] Kass M, Witkin A, Terzopoulos D. Snakes: Active contour models[J]. InternationalJournal of Computer Vision,1988,1(4):321-331
[74] Rogers M, Graham J. Robust active shape model search[J]. Lecture Notes in ComputerScience,2002,2(3):289-312
[75] Cootes T F, Edwards G J, Taylor C J. Active appearance models[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,2001,23(6):681-685
[76] Stegmann M B. Active appearance models: Theory, extensions and cases[M].Copenhagen: Technical University of Denmark,2000:50-53
[77] Gao X, Su Y, Li X, et al. A review of active appearance models[J]. IEEE Transactionson Systems, Man, and Cybernetics,2010,40(2):145-158
[78] Ghiass R S, Arandjelovic O, Bendada H, et al. Illumination-invariant face recognitionfrom a single image across extreme pose using a dual dimension AAM ensemble in thethermal infrared spectrum[A]. Proceedings of International Joint Conference on NeuralNetworks[C]. IEEE,2013:1-10
[79] Zhou L, Fang B, Li W, et al. Improved active shape model for facial feature localizationusing poem descriptor[A]. Proceedings of International Conference on Wavelet Analysisand Pattern Recognition[C]. IEEE,2013:184-189
[80] Zaker N, Mahoor M H, Mattson W I, et al. A comparison of alternative classifiers fordetecting occurrence and intensity in spontaneous facial expression of infants with theirmothers[A]. Proceedings of10th IEEE International Conference and Workshops onAutomatic Face and Gesture Recognition[C]. IEEE,2013:1-6
[81]Hachisuka S. Human and Vehicle-Driver Drowsiness Detection by Facial Expression[A].Proceedings of International Conference on Biometrics and Kansei Engineering[C].IEEE,2013:320-326
[82] Mitchell S C, Bosch J G, Lelieveldt B P F, et al.3-D active appearance models:segmentation of cardiac MR and ultrasound images[J]. IEEE Transactions on MedicalImaging,2002,21(9):1167-1178
[83] Wu Y, Wang Z, Ji Q. Facial Feature Tracking under Varying Facial Expressions andFace Poses based on Restricted Boltzmann Machines[A]. Proceedings of IEEEConference on Computer Vision and Pattern Recognition[C]. IEEE,2013:3452-3459
[84] Sung J, Kanade T, Kim D. Pose robust face tracking by combining active appearancemodels and cylinder head models[J]. International Journal of Computer Vision,2008,80(2):260-274
[85] Black M J, Jepson A D. Eigentracking: Robust matching and tracking of articulatedobjects using a view-based representation[J]. International Journal of Computer Vision,1998,26(1):63-84
[86] Ross D A, Lim J, Lin R S, et al. Incremental learning for robust visual tracking[J].International Journal of Computer Vision,2008,77(1):125-141
[87] Li X, Hu W, Zhang Z, et al. Visual tracking via incremental log-Euclidean Riemanniansubspace learning[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C]. Anchorage, AK: IEEE,2008:1-8
[88] Candes E J, Li X, Ma Y, et al. Robust principal component analysis?[J]. Journal of theACM,2011,58(3):1-37
[89] Ho J, Lee K C, Yang M H, et al. Visual tracking using learned linear subspaces[A].Proceedings of2004IEEE Computer Society Conference on Computer Vision andPattern Recognition[C]. IEEE,2004,1:782-789
[90] Lin D, Zheng H, Ma D. Robust visual tracking using local salient coding and PCAsubspace modeling[A]. Proceedings of IEEE International Workshop on InformationForensics and Security[C]. Guangzhou: IEEE,2013:25-30
[91] Lowe D G. Distinctive image features from scale-invariant keypoints[J]. Internationaljournal of computer vision,2004,60(2):91-110
[92] Narayanan V K, Crane C D. Active relearning for robust on-road vehicle detection andtracking[A]. Proceedings of the13th International Conference on Control, Automationand Systems[C]. IEEE,2013:124-129
[93] Yang P, Shan S, Gao W, et al. Face recognition using ada-boosted gabor features[A].Proceedings of the6th IEEE International Conference on Automatic Face and GestureRecognition[C]. IEEE,2004:356-361
[94] Zoidi O, Tefas A, Pitas I. Visual object tracking via Gabor-based salient featuresextraction[A]. Proceedings of the20th European Signal Processing Conference[C]. IEEE,2012:1925-1929
[95] Hsieh J-W, Chen L-C, Chen D-Y. Symmetrical SURF and Its Applications to VehicleDetection and Vehicle Make and Model Recognition[J]. IEEE Transactions onIntelligent Transportation Systems,2014,15(1):6-20
[96] Bay H, Ess A, Tuytelaars T, et al. Speeded-up robust features (SURF)[J]. ComputerVision and Image Understanding,2008,110(3):346-359
[97] Tola E, Lepetit V, Fua P. Daisy: An efficient dense descriptor applied to wide-baselinestereo[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(5):815-830
[98] Wright J, Yang A Y, Ganesh A, et al. Robust face recognition via sparserepresentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(2):210-227
[99] Mei X, Ling H. Robust visual tracking using l1minimization[A]. Proceedings of the12th International Conference on Computer Vision[C]. IEEE,2009:1436-1443
[100] Chen F, Wang Q, Wang S, et al. Object tracking via appearance modeling and sparserepresentation[J]. Image and Vision Computing,2011,29:787-796
[101] Kuang J, Zhou X, Gamst A. Robust Visual Cooperative Tracking Using ConstrainedAdaptive Sparse Representations and Sparse Classifier Grids[J]. IEEE Transactions onCircuits and Systems for Video Technology,2014, PP (99):1-15
[102] Bai T, Li Y. Robust visual tracking using flexible structured sparse representation[J].IEEE Transactions on Industrial Informatics,2014,10(1):538-547
[103] Video Surveillance Online Repository[DB/OL]. http://www.openvisor.org
[104] Fraley C, Raftery A E,. How many clusters? Which clustering method? Answers viamodel-based cluster analysis[J]. The computer journal,1998,41(8):578-588
[105] Konishi S, Kitagawa G. Information criteria and statistical modeling[M]. Springer,2007:5-9
[106] Zivkovic Z, van der Heijden F. Recursive Unsupervised Learning of Finite MixtureModels[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(5):651-656
[107] Biernacki C, Celeux G, Govaert G. Choosing starting values for the EM algorithmfor getting the highest likelihood in multivariate gaussian mixture models[J].Computational Statistics&Data Analysis,2003,41(3):561-575
[108]李全民,张运楚.自适应混合高斯模型的改进设计[J].计算机应用,2007,27(8):2014-2017
[109]陈祖爵,陈潇君,何鸿.基于改进的混合高斯模型的运动目标检测[J].中国图象图形学报,2007,12(9):1585-1589
[110]苗东升.论复杂性[J].自然辩证法通讯,2000,22(6):87-92
[111]张学文.组成论[M].合肥:中国科学技术大学出版社,2003:19-87
[112] Performance Evaluation in Tracking for Surveillance (PETS)[DB/OL].http://ftp.pets.rdg.ac.uk
[113] Black J, Ellis T, Rosin P. A novel method for video tracking performance evaluation[A]. Proceedings of Joint IEEE International Workshop on Visual Surveillance andPerformance Evaluation of Tracking and Surveillance [C]. Nice, France: IEEE,2003:125-132
[114]李鹏飞,陈朝武,李晓峰.智能视频算法评估综述[J].计算机辅助设计与图形学学报,2010,22(2):352-360
[115] Hojjatoleslami S, Kittler J. Region growing: A new approach[J]. IEEE Transactionson Image Processing,1998,7(7):1079-1084
[116] Gan G, Ma C, Wu J. Data Clustering: Theory, Algorithms, and Applications[M].Asa-Siam, USA:2007:22-35
[117] Kanungo T, Mount D M, Netanyahu N S, et al. An efficient k-means clusteringalgorithm: Analysis and implementation[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,2002,24(7):881-892
[118] Cannon R L, Dave J V, Bezdek J C. Efficient implementation of the fuzzy c-meansclustering algorithms[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1998,8(2):248-255
[119] Yager R, Filev D. Generation of fuzzy rules by mountain clustering[J]. Journal ofIntelligent and Fuzzy Systems,1994,2(3):209-219
[120] Chiu S L. Fuzzy model identification based on cluster estimation[J]. Journal ofIntelligent and Fuzzy Systems,1994,2(3):267-278
[121] Theodoridis S, Koutroumbas K. Pattern Recognition[M].4ed. London: AcademicPress,2008:595-620
[122] Jain A K, Murty M N, Flynn P J. Data clustering: a review[J]. ACM computingsurveys (CSUR),1999,31(3):264-323
[123] Yang T, Li S Z, Pan Q, et al. Real-time and accurate segmentation of moving objectsin dynamic scene[A]. Proceedings of the ACM2nd International Workshop on VideoSurveillance&Sensor Networks[C]. ACM New York,2004:136-143
[124]周均扬.贝叶斯动态线性模型介绍及常量模型分析[D].广州:中山大學,2003
[125] Chen Z. Bayesian filtering: from Kalman filters to Particle filters, and beyond[J].ACM computing surveys,2003,31(3):2-14
[126] Chien Y, Fu K. On Bayesian learning and stochastic approximation[J]. IEEETransactions on Systems Science and Cybernetics,1967,3(1):28-38
[127]林元烈.应用随机过程[M].北京:清华大学出版社,2002:78-82
[128] Van Der Merwe R. Sigma-point Kalman filters for probabilistic inference in dynamicstate-space models[D]. PhD thesis, University of Stellenbosch,2004
[129] Haykin S S. Kalman filtering and neural networks[M]. New York: Wiley,2001:221-280
[130] Julier S, Uhlmann J, Durrant-Whyte H F. A new method for the nonlineartransformation of means and covariances in filters and estimators[J]. IEEE Transactionson Automatic Control,2000,45(3):477-482
[131] Julier S J. The scaled unscented transformation[A]. Proceedings of the2002American Control Conference[C]. IEEE,2002,6:4555-4559
[132] Solomon H. Buffon needle problem, extensions, and estimation of π[M].Philadelphia,PA: SIAM,1978:1-24
[133] Von Neumann J. Various techniques used in connection with random digits[J].Applied Math Series,1951,12(36-38):1
[134] Doucet A, Godsill S, Andrieu C. On sequential Monte Carlo sampling methods forBayesian filtering[J]. Statistics and computing,2000,10(3):197-208
[135] Liu J S, Chen R. Blind deconvolution via sequential imputations[J]. Journal of theAmerican Statistical Association,1995,90(430):567-576
[136] Hol J D, Schon T B, Gustafsson F. On resampling algorithms for particle filters[A].Proceedings of IEEE Nonlinear Statistical Signal Processing Workshop[C]. IEEE,2006:79-82
[137]齐欢.系统建模与仿真[M].北京:清华大学出版社,2004:213
[138] Cheng Y. Mean shift, mode seeking, and clustering [J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,1995,17(8):790-799
[139] Comaniciu D, Meer P. Mean shift analysis and applications[A]. Proceedings of the7th IEEE International Conference on Computer Vision[C]. IEEE,1999,2:1197-1203
[140] Comaniciu D, Meer P. Mean shift: A robust approach toward feature space analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603-619
[141] Visual Tracking Decomposition dataset[DB/OL]. http://cv.snu.ac.kr/research/~vtd
[142] Xu C, Prince J L. Snakes, shapes, and gradient vector flow[J]. IEEE Transactions onImage Processing,1998,7(3):359-369
[143] Li B, Acton S T. Active contour external force using vector field convolution forimage segmentation[J]. IEEE Transactions on Image Processing,2007,16(8):2096-2106
[144] Caselles V, CattéF, Coll T, et al. A geometric model for active contours in imageprocessing[J]. Numerische mathematik,1993,66(1):1-31
[145] Caselles V, Kimmel R, Sapiro G. Geodesic active contours[J]. International Journalof Computer Vision,1997,22(1):61-79
[146] Sethian J A. Level set methods and fast marching methods[J]. Journal of Computingand Information Technology,2003,11(1):1-2
[147] Zhao H-K, Chan T, Merriman B, et al. A variational level set approach to multiphasemotion[J]. Journal of computational physics,1996,127(1):179-195
[148] Chan T F, Vese L A. Active contours without edges[J]. IEEE Transactions on ImageProcessing,2001,10(2):266-277
[149] Peng D, Merriman B, Osher S, et al. A PDE-based fast local level set method[J].Journal of computational physics,1999,155(2):410-438
[150] Gomes J, Faugeras O. Reconciling distance functions and level sets[A]. Proceedingsof the5th IEEE EMBS International Summer School on Biomedical Imaging[C]. IEEE,2002:15
[151] Li C, Xu C, Gui C, et al. Distance regularized level set evolution and its applicationto image segmentation[J]. IEEE Transactions on Image Processing,2010,19(12):3243-3254
[152] Chung D H, Sapiro G. On the level lines and geometry of vector-valued images[J].IEEE Signal Processing Letters,2000,7(9):241-243
[153] Otsu N. A threshold selection method from gray-level histograms[J]. Automatica,1975,11(285-296):23-27
[154] Li C, Xu C, Gui C, et al. Level set evolution without re-initialization: a newvariational formulation[A]. Proceedings of IEEE Computer Society Conference onComputer Vision and Pattern Recognition[C]. IEEE,2005,1:430-436
[155]吴小庆.数学物理方程及其应用[M].北京:科学出版社,2008:14-19
[156] CASIA Gait Database[DB/OL]. http://www.sinobiometrics.com
[2] Collins R T, Lipton A J, Kanade T A System for Video Surveillance and Monitoring:VSAM final report[R]. Robotic Institute Carnegie Mellon University,2000
[3] Remagnino P, Tan T, Baker K. Multi-agent visual surveillance of dynamic scenes[J].Image and Vision Computing,1998,16(8):529-532
[4] Maggioni C, K mmerer B, Gesturecomputer－history, design and applications[A].Cippola R, Pentland A. In Computer Vision for Human-Machine Interaction[M],Cambridge University Press: Cambridge,1998:23-52
[5] Freeman W T, Weissman C. Television control by hand gestures[A]. Proceedings ofInternational Workshop on Automatic Face and Gesture Recognition[C].1995:179-183
[6] Haritaoglu I, Harwood D, Davis L S. W4: real-time surveillance of people and theiractivities[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(8):809-830
[7] Pavlidis I, Morellas V, Tsiamyrtzis P, et al. Urban surveillance systems: from thelaboratory to the commercial world[J]. Proceedings of the IEEE,2001,89(10):1478-1497
[8] Bogaert M, Chelq N, Cornez P, et al. The PASSWORDS Project (intelligent video imageanalysis system)[A]. Proceedings of International Conference on Image Processing[C].IEEE,1996,3:675-678
[9] Wren C R, Azarbayejani A, Darrell T, et al. Pfinder: Real-time tracking of the humanbody[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19(7):780-785
[10] Coifman B, Beymer D, McLauchlan P, et al. A real-time computer vision system forvehicle tracking and traffic surveillance[J]. Transportation Research Part C: EmergingTechnologies,1998,6(4):271-288
[11] Siebel N T, Maybank S. The ADVISOR visual surveillance system[A]. Proceedings ofthe ECCV2004workshop Applications of Computer Vision[C]. Prague, Czech Republic:2004,1:103-111
[12] Reardon S, FBI launches$1billion face recognition project [N]. New Scientist.7September,2012,2880:20
[13] Arif M, Samani H, Yang C-Y, et al. Adaptation of mobile robots to intelligentvehicles[A]. Proceedings of the4th IEEE International Conference on SoftwareEngineering and Service Science [C]. Beijing: IEEE,2013:550-553
[14] Jiang G-Q, Zhao C-J. Apple recognition based on machine vision[A]. Proceedings ofInternational Conference on Machine Learning and Cybernetics[C]. IEEE,2012,3:1148-1151
[15] Jiang G-Q, Zhao C-J, Si Y-S. A machine vision based crop rows detection foragricultural robots[A]. Proceedings of International Conference on Wavelet Analysis andPattern Recognition[C]. IEEE,2010:114-118
[16] Yesin K, Nelson B J, Papanikolopoulos N P, et al. Active video system for a miniaturereconnaissance robot[A]. Proceedings of IEEE International Conference on Robotics andAutomation[C]. IEEE,2000,4:3919-3924
[17] Wu Y, Huang T. Vision-based gesture recognition: A review[J]. Gesture-basedcommunication in human-computer interaction,1999:103-115
[18] Koh E, Won J, Bae C. Vision-based Virtual Touch Screen Interface[A]. Proceedings ofInternational Conference on Consumer Electronics[C]. IEEE,2008:1-2
[19] Pavlovic V I, Sharma R, Huang T S. Visual interpretation of hand gestures forhuman-computer interaction: A review[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,1997,19(7):677-695
[20] Kwok K W, Yam Y, Lo K W. Vision system and projective rectification for a robotdrawing platform[A]. Proceedings of International Conference on Control andAutomation[C]. IEEE,2005,2:691-696
[21] Zhang J, Sui L, Zhuo L, et al. Pornographic image region detection based on visualattention model in compressed domain[J]. IET Image Processing,2013,7(4):384-391
[22] Neuman M R, Baura G D, Meldrum S, et al. Advances in medical devices and medicalelectronics[J]. Proceedings of the IEEE,2012,100(Special Centennial Issue):1537-1550
[23] Kaiser M, Gans N, Dixon W. Vision-based estimation for guidance, navigation, andcontrol of an aerial vehicle[J]. IEEE Transactions on Aerospace and Electronic Systems,2010,46(3):1064-1077
[24] Huang H-B, Huo H, Fang T. Hierarchical Manifold Learning With Applications toSupervised Classification for High-Resolution Remotely Sensed Images[J]. IEEETransactions on Geoscience and Remote Sensing,2014,52(3):1677-1692
[25] Stauffer C, Grimson W. Adaptive background mixture models for real-time tracking[A].Proceedings of IEEE Computer Society Conference on Computer Vision and PatternRecognition[C]. Fort Collins: IEEE,1999:246-252
[26] Power P W, Schoonees J A. Understanding background mixture models for foregroundsegmentation[A]. Proceedings of Image and Vision Computing[C]. New Zealand:University of Auckland,2002,2002:267-271
[27] KaewTraKulPong P, Bowden R, An improved adaptive background mixture model forreal-time tracking with shadow detection[A]. Jones G A, Paragios N, Regazzoni C S. InVideo-Based Surveillance Systems[M], Springer US:2001,2:135-144
[28] Lee K C, Ho J, Yang M H, et al. Visual tracking and recognition using probabilisticappearance manifolds[J]. Computer Vision and Image Understanding,2005,99(3):303-331
[29] Zivkovic Z. Improved adaptive Gaussian mixture model for background subtraction[A].Proceedings of the17th International Conference on Pattern Recognition[C]. IEEE,2004,2:28-31
[30] Li D, Xu L, Goodman E. Online background learning for illumination-robust foregrounddetection[A]. Proceedings of the11th International Conference on Control AutomationRobotics&Vision[C]. Singapore: IEEE,2010:1093-1100
[31] Bouttefroy P L M, Bouzerdoum A, Phung S L, et al. On the analysis of backgroundsubtraction techniques using Gaussian mixture models[A]. Proceedings of IEEEInternational Conference on Acoustics Speech and Signal Processing[C]. IEEE,2010:4042-4045
[32] Huang T, Qiu J, Sakayori T, et al. Motion detection based on background modeling andperformance analysis for outdoor surveillance[A]. Proceedings of InternationalConference on Computer Modeling and Simulation[C]. IEEE,2009:38-42
[33] Liu Y, Bin Z. The improved moving object detection and shadow removing algorithmsfor video surveillance[A]. Proceedings of International Conference on ComputationalIntelligence and Software Engineering[C]. IEEE,2010:1-5
[34] Lin H-H, Chuang J-H, Liu T-L. Regularized background adaptation: a novel learningrate control scheme for Gaussian mixture modeling[J]. IEEE Transactions on ImageProcessing,2011,20(3):822-836
[35] Elgammal A, Duraiswami R, Harwood D, et al. Background and foreground modelingusing nonparametric kernel density estimation for visual surveillance[J]. Proceedings ofthe IEEE,2002,90(7):1151-1163
[36] Kim K, Chalidabhongse T H, Harwood D, et al. Real-time foreground–backgroundsegmentation using codebook model[J]. Real-time imaging,2005,11(3):172-185
[37] Zhang Z H, Chen R Q, Lu H Q, et al. Moving Foreground Detection Based on ModifiedCodebook[A]. Proceedings of the2nd International Congress on Image and SignalProcessing[C]. IEEE,2009:1-5
[38] Prati A, Mikic I, Trivedi M M, et al. Detecting moving shadows: algorithms andevaluation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(7):918-923
[39] Gibson J J. The perception of the visual world[M]. Houghton Mifflin, England: Oxford,1950:37-91
[40] Horn B K P. Robot vision[M]. MIT Press,1986:278-280
[41] Horn B K P, Schunck B G. Determining optical flow[J]. Artificial intelligence,1981,17(1):185-203
[42] Barron J L, Fleet D J, Beauchemin S S. Performance of optical flow techniques[J].International Journal of Computer Vision,1994,12(1):43-77
[43] Opelt A, Pinz A, Zisserman A. A boundary-fragment-model for object detection[A].Leonardis A, Bischof H, Pinz A. Proceedings of the9th European Conference onComputer Vision[C]. Graz, Austria: Springer Berlin Heidelberg,2006,2:575-588
[44] Dorkó G, Schmid C. Selection of scale-invariant parts for object class recognition[A].Proceedings of the9th IEEE International Conference on Computer Vision[C]. IEEE,2013:634-639
[45] Haralick R M, Shanmugam K, Dinstein I H. Textural features for image classification[J].IEEE Transactions on Systems, Man and Cybernetics,1973,(6):610-621
[46]高隽,谢昭.图像理解理论与方法[M].北京:科学出版社,2009:179-186
[47] Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,2003,25(5):564-577
[48] Hager G D, Belhumeur P N. Efficient region tracking with parametric models ofgeometry and illumination[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1998,20(10):1025-1039
[49] Azarbayejani A, Pentland A P. Recursive estimation of motion, structure, and focallength[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995,17(6):562-575
[50] Isard M, Blake A. Condensation: conditional density propagation for visual tracking[J].International Journal of Computer Vision,1998,29(1):5-28
[51] Han B, Davis L. On-line density-based appearance modeling for object tracking[A].Proceedings of the10th IEEE International Conference on Computer Vision[C]. IEEE,2005,2:1492-1499
[52] Wang H, Suter D, Schindler K, et al. Adaptive object tracking based on an effectiveappearance filter[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(9):1661-1667
[53] Pawar M. Mean shift Face Tracking with Dynamic Target Model update using BayesianSkin Classifier[A]. Proceedings of International Conference on ComputationalIntelligence&Computing Research[C]. Coimbatore: IEEE2012:1-5
[54] Fieguth P, Terzopoulos D. Color-based tracking of heads and other mobile objects atvideo frame rates[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C]. IEEE,1997:21-27
[55] Zoidi O, Tefas A, Pitas I. Visual object tracking based on local steering kernels and colorhistograms[J]. IEEE Transactions on Circuits and Systems for Video Technology,2013,23(5):870-882
[56] Bradski G R. Computer vision face tracking for use in a perceptual user interface[J].Intel Technology Journal Q2,1998,2(2):1-15
[57] Comaniciu D, Ramesh V, Meer P. Real-time tracking of non-rigid objects using meanshift[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C]. IEEE,2000,2:142-149
[58] Pérez P, Hue C, Vermaak J, et al., Color-based probabilistic tracking[A]. In the7thEuropean Conference on Computer Vision[M], Springer: Berlin Heidelberg,2002:661-675
[59] Collins R T. Mean-shift blob tracking through scale space[A]. Proceedings of IEEEComputer Society Conference on Computer Vision and Pattern Recognition[C]. IEEE,2003,2:234-240
[60] Adam A, Rivlin E, Shimshoni I. Robust fragments-based tracking using the integralhistogram[A]. Proceedings of IEEE Computer Society Conference on Computer Visionand Pattern Recognition[C]. IEEE,2006,1:798-805
[61] Stern H, Efros B. Adaptive color space switching for face tracking in multi-coloredlighting environments[A]. Proceedings of the5th IEEE International Conference onAutomatic Face and Gesture Recognition[C]. IEEE,2002:249-254
[62] Collins R T, Liu Y, Leordeanu M. Online selection of discriminative tracking features[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(10):1631-1643
[63] Dalal N, Triggs B. Histograms of oriented gradients for human detection[A].Proceedings of Computer Society Conference on Computer Vision and PatternRecognition[C]. IEEE,2005,1:886-893
[64] Zhu W, Levinson S E. Edge orientation-based multi-view object recognition[A].Proceedings of the15th International Conference on Pattern Recognition[C]. Barcelona:IEEE,2000,2000:1936-1939
[65] Yao B, Liu Z, Nie X, et al. Animated Pose Templates for Modelling and DetectingHuman Actions[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2014,36(3):436-452
[66] Jia W, Hu R-X, Lei Y-K, et al. Histogram of Oriented Lines for PalmprintRecognition[J]. IEEE Transactions on Systems, Man, and Cybernetics,2014,44(3):385-395
[67] Agarwal S, Verma A, Singh P. Content Based Image Retrieval using Discrete WaveletTransform and Edge Histogram Descriptor[A]. Proceedings of International Conferenceon Information Systems and Computer Networks[C]. IEEE,2013:19-23
[68] Lucas B D, Kanade T. An iterative image registration technique with an application tostereo vision[A]. Proceedings of the7th International Joint Conference on ArtificialIntelligence[C].1981:674-679
[69] Frey B J. Filling in scenes by propagating probabilities through layers and intoappearance models[A]. Proceedings of IEEE Conference on Computer Vision andPattern Recognition[C]. IEEE,2000,1:185-192
[70] Olson C F. Maximum-likelihood template matching[A]. Proceedings of IEEEConference on Computer Vision and Pattern Recognition[C]. Hilton Head Island, SC:IEEE,2000,2:52-57
[71] McInerney T, Terzopoulos D. Deformable models in medical image analysis[A].Proceedings of the Workshop on Mathematical Methods in Biomedical ImageAnalysis[C]. IEEE,1996:171-180
[72] Fisker R. Making deformable template models operational[D]. PhD thesis, Copenhagen:Informatics and Mathematical Modelling, Technical University of Denmark,2000
[73] Kass M, Witkin A, Terzopoulos D. Snakes: Active contour models[J]. InternationalJournal of Computer Vision,1988,1(4):321-331
[74] Rogers M, Graham J. Robust active shape model search[J]. Lecture Notes in ComputerScience,2002,2(3):289-312
[75] Cootes T F, Edwards G J, Taylor C J. Active appearance models[J]. IEEE Transactionson Pattern Analysis and Machine Intelligence,2001,23(6):681-685
[76] Stegmann M B. Active appearance models: Theory, extensions and cases[M].Copenhagen: Technical University of Denmark,2000:50-53
[77] Gao X, Su Y, Li X, et al. A review of active appearance models[J]. IEEE Transactionson Systems, Man, and Cybernetics,2010,40(2):145-158
[78] Ghiass R S, Arandjelovic O, Bendada H, et al. Illumination-invariant face recognitionfrom a single image across extreme pose using a dual dimension AAM ensemble in thethermal infrared spectrum[A]. Proceedings of International Joint Conference on NeuralNetworks[C]. IEEE,2013:1-10
[79] Zhou L, Fang B, Li W, et al. Improved active shape model for facial feature localizationusing poem descriptor[A]. Proceedings of International Conference on Wavelet Analysisand Pattern Recognition[C]. IEEE,2013:184-189
[80] Zaker N, Mahoor M H, Mattson W I, et al. A comparison of alternative classifiers fordetecting occurrence and intensity in spontaneous facial expression of infants with theirmothers[A]. Proceedings of10th IEEE International Conference and Workshops onAutomatic Face and Gesture Recognition[C]. IEEE,2013:1-6
[81]Hachisuka S. Human and Vehicle-Driver Drowsiness Detection by Facial Expression[A].Proceedings of International Conference on Biometrics and Kansei Engineering[C].IEEE,2013:320-326
[82] Mitchell S C, Bosch J G, Lelieveldt B P F, et al.3-D active appearance models:segmentation of cardiac MR and ultrasound images[J]. IEEE Transactions on MedicalImaging,2002,21(9):1167-1178
[83] Wu Y, Wang Z, Ji Q. Facial Feature Tracking under Varying Facial Expressions andFace Poses based on Restricted Boltzmann Machines[A]. Proceedings of IEEEConference on Computer Vision and Pattern Recognition[C]. IEEE,2013:3452-3459
[84] Sung J, Kanade T, Kim D. Pose robust face tracking by combining active appearancemodels and cylinder head models[J]. International Journal of Computer Vision,2008,80(2):260-274
[85] Black M J, Jepson A D. Eigentracking: Robust matching and tracking of articulatedobjects using a view-based representation[J]. International Journal of Computer Vision,1998,26(1):63-84
[86] Ross D A, Lim J, Lin R S, et al. Incremental learning for robust visual tracking[J].International Journal of Computer Vision,2008,77(1):125-141
[87] Li X, Hu W, Zhang Z, et al. Visual tracking via incremental log-Euclidean Riemanniansubspace learning[A]. Proceedings of IEEE Conference on Computer Vision and PatternRecognition[C]. Anchorage, AK: IEEE,2008:1-8
[88] Candes E J, Li X, Ma Y, et al. Robust principal component analysis?[J]. Journal of theACM,2011,58(3):1-37
[89] Ho J, Lee K C, Yang M H, et al. Visual tracking using learned linear subspaces[A].Proceedings of2004IEEE Computer Society Conference on Computer Vision andPattern Recognition[C]. IEEE,2004,1:782-789
[90] Lin D, Zheng H, Ma D. Robust visual tracking using local salient coding and PCAsubspace modeling[A]. Proceedings of IEEE International Workshop on InformationForensics and Security[C]. Guangzhou: IEEE,2013:25-30
[91] Lowe D G. Distinctive image features from scale-invariant keypoints[J]. Internationaljournal of computer vision,2004,60(2):91-110
[92] Narayanan V K, Crane C D. Active relearning for robust on-road vehicle detection andtracking[A]. Proceedings of the13th International Conference on Control, Automationand Systems[C]. IEEE,2013:124-129
[93] Yang P, Shan S, Gao W, et al. Face recognition using ada-boosted gabor features[A].Proceedings of the6th IEEE International Conference on Automatic Face and GestureRecognition[C]. IEEE,2004:356-361
[94] Zoidi O, Tefas A, Pitas I. Visual object tracking via Gabor-based salient featuresextraction[A]. Proceedings of the20th European Signal Processing Conference[C]. IEEE,2012:1925-1929
[95] Hsieh J-W, Chen L-C, Chen D-Y. Symmetrical SURF and Its Applications to VehicleDetection and Vehicle Make and Model Recognition[J]. IEEE Transactions onIntelligent Transportation Systems,2014,15(1):6-20
[96] Bay H, Ess A, Tuytelaars T, et al. Speeded-up robust features (SURF)[J]. ComputerVision and Image Understanding,2008,110(3):346-359
[97] Tola E, Lepetit V, Fua P. Daisy: An efficient dense descriptor applied to wide-baselinestereo[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(5):815-830
[98] Wright J, Yang A Y, Ganesh A, et al. Robust face recognition via sparserepresentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(2):210-227
[99] Mei X, Ling H. Robust visual tracking using l1minimization[A]. Proceedings of the12th International Conference on Computer Vision[C]. IEEE,2009:1436-1443
[100] Chen F, Wang Q, Wang S, et al. Object tracking via appearance modeling and sparserepresentation[J]. Image and Vision Computing,2011,29:787-796
[101] Kuang J, Zhou X, Gamst A. Robust Visual Cooperative Tracking Using ConstrainedAdaptive Sparse Representations and Sparse Classifier Grids[J]. IEEE Transactions onCircuits and Systems for Video Technology,2014, PP (99):1-15
[102] Bai T, Li Y. Robust visual tracking using flexible structured sparse representation[J].IEEE Transactions on Industrial Informatics,2014,10(1):538-547
[103] Video Surveillance Online Repository[DB/OL]. http://www.openvisor.org
[104] Fraley C, Raftery A E,. How many clusters? Which clustering method? Answers viamodel-based cluster analysis[J]. The computer journal,1998,41(8):578-588
[105] Konishi S, Kitagawa G. Information criteria and statistical modeling[M]. Springer,2007:5-9
[106] Zivkovic Z, van der Heijden F. Recursive Unsupervised Learning of Finite MixtureModels[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(5):651-656
[107] Biernacki C, Celeux G, Govaert G. Choosing starting values for the EM algorithmfor getting the highest likelihood in multivariate gaussian mixture models[J].Computational Statistics&Data Analysis,2003,41(3):561-575
[108]李全民,张运楚.自适应混合高斯模型的改进设计[J].计算机应用,2007,27(8):2014-2017
[109]陈祖爵,陈潇君,何鸿.基于改进的混合高斯模型的运动目标检测[J].中国图象图形学报,2007,12(9):1585-1589
[110]苗东升.论复杂性[J].自然辩证法通讯,2000,22(6):87-92
[111]张学文.组成论[M].合肥:中国科学技术大学出版社,2003:19-87
[112] Performance Evaluation in Tracking for Surveillance (PETS)[DB/OL].http://ftp.pets.rdg.ac.uk
[113] Black J, Ellis T, Rosin P. A novel method for video tracking performance evaluation[A]. Proceedings of Joint IEEE International Workshop on Visual Surveillance andPerformance Evaluation of Tracking and Surveillance [C]. Nice, France: IEEE,2003:125-132
[114]李鹏飞,陈朝武,李晓峰.智能视频算法评估综述[J].计算机辅助设计与图形学学报,2010,22(2):352-360
[115] Hojjatoleslami S, Kittler J. Region growing: A new approach[J]. IEEE Transactionson Image Processing,1998,7(7):1079-1084
[116] Gan G, Ma C, Wu J. Data Clustering: Theory, Algorithms, and Applications[M].Asa-Siam, USA:2007:22-35
[117] Kanungo T, Mount D M, Netanyahu N S, et al. An efficient k-means clusteringalgorithm: Analysis and implementation[J]. IEEE Transactions on Pattern Analysis andMachine Intelligence,2002,24(7):881-892
[118] Cannon R L, Dave J V, Bezdek J C. Efficient implementation of the fuzzy c-meansclustering algorithms[J]. IEEE Transactions on Pattern Analysis and MachineIntelligence,1998,8(2):248-255
[119] Yager R, Filev D. Generation of fuzzy rules by mountain clustering[J]. Journal ofIntelligent and Fuzzy Systems,1994,2(3):209-219
[120] Chiu S L. Fuzzy model identification based on cluster estimation[J]. Journal ofIntelligent and Fuzzy Systems,1994,2(3):267-278
[121] Theodoridis S, Koutroumbas K. Pattern Recognition[M].4ed. London: AcademicPress,2008:595-620
[122] Jain A K, Murty M N, Flynn P J. Data clustering: a review[J]. ACM computingsurveys (CSUR),1999,31(3):264-323
[123] Yang T, Li S Z, Pan Q, et al. Real-time and accurate segmentation of moving objectsin dynamic scene[A]. Proceedings of the ACM2nd International Workshop on VideoSurveillance&Sensor Networks[C]. ACM New York,2004:136-143
[124]周均扬.贝叶斯动态线性模型介绍及常量模型分析[D].广州:中山大學,2003
[125] Chen Z. Bayesian filtering: from Kalman filters to Particle filters, and beyond[J].ACM computing surveys,2003,31(3):2-14
[126] Chien Y, Fu K. On Bayesian learning and stochastic approximation[J]. IEEETransactions on Systems Science and Cybernetics,1967,3(1):28-38
[127]林元烈.应用随机过程[M].北京:清华大学出版社,2002:78-82
[128] Van Der Merwe R. Sigma-point Kalman filters for probabilistic inference in dynamicstate-space models[D]. PhD thesis, University of Stellenbosch,2004
[129] Haykin S S. Kalman filtering and neural networks[M]. New York: Wiley,2001:221-280
[130] Julier S, Uhlmann J, Durrant-Whyte H F. A new method for the nonlineartransformation of means and covariances in filters and estimators[J]. IEEE Transactionson Automatic Control,2000,45(3):477-482
[131] Julier S J. The scaled unscented transformation[A]. Proceedings of the2002American Control Conference[C]. IEEE,2002,6:4555-4559
[132] Solomon H. Buffon needle problem, extensions, and estimation of π[M].Philadelphia,PA: SIAM,1978:1-24
[133] Von Neumann J. Various techniques used in connection with random digits[J].Applied Math Series,1951,12(36-38):1
[134] Doucet A, Godsill S, Andrieu C. On sequential Monte Carlo sampling methods forBayesian filtering[J]. Statistics and computing,2000,10(3):197-208
[135] Liu J S, Chen R. Blind deconvolution via sequential imputations[J]. Journal of theAmerican Statistical Association,1995,90(430):567-576
[136] Hol J D, Schon T B, Gustafsson F. On resampling algorithms for particle filters[A].Proceedings of IEEE Nonlinear Statistical Signal Processing Workshop[C]. IEEE,2006:79-82
[137]齐欢.系统建模与仿真[M].北京:清华大学出版社,2004:213
[138] Cheng Y. Mean shift, mode seeking, and clustering [J]. IEEE Transactions on PatternAnalysis and Machine Intelligence,1995,17(8):790-799
[139] Comaniciu D, Meer P. Mean shift analysis and applications[A]. Proceedings of the7th IEEE International Conference on Computer Vision[C]. IEEE,1999,2:1197-1203
[140] Comaniciu D, Meer P. Mean shift: A robust approach toward feature space analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603-619
[141] Visual Tracking Decomposition dataset[DB/OL]. http://cv.snu.ac.kr/research/~vtd
[142] Xu C, Prince J L. Snakes, shapes, and gradient vector flow[J]. IEEE Transactions onImage Processing,1998,7(3):359-369
[143] Li B, Acton S T. Active contour external force using vector field convolution forimage segmentation[J]. IEEE Transactions on Image Processing,2007,16(8):2096-2106
[144] Caselles V, CattéF, Coll T, et al. A geometric model for active contours in imageprocessing[J]. Numerische mathematik,1993,66(1):1-31
[145] Caselles V, Kimmel R, Sapiro G. Geodesic active contours[J]. International Journalof Computer Vision,1997,22(1):61-79
[146] Sethian J A. Level set methods and fast marching methods[J]. Journal of Computingand Information Technology,2003,11(1):1-2
[147] Zhao H-K, Chan T, Merriman B, et al. A variational level set approach to multiphasemotion[J]. Journal of computational physics,1996,127(1):179-195
[148] Chan T F, Vese L A. Active contours without edges[J]. IEEE Transactions on ImageProcessing,2001,10(2):266-277
[149] Peng D, Merriman B, Osher S, et al. A PDE-based fast local level set method[J].Journal of computational physics,1999,155(2):410-438
[150] Gomes J, Faugeras O. Reconciling distance functions and level sets[A]. Proceedingsof the5th IEEE EMBS International Summer School on Biomedical Imaging[C]. IEEE,2002:15
[151] Li C, Xu C, Gui C, et al. Distance regularized level set evolution and its applicationto image segmentation[J]. IEEE Transactions on Image Processing,2010,19(12):3243-3254
[152] Chung D H, Sapiro G. On the level lines and geometry of vector-valued images[J].IEEE Signal Processing Letters,2000,7(9):241-243
[153] Otsu N. A threshold selection method from gray-level histograms[J]. Automatica,1975,11(285-296):23-27
[154] Li C, Xu C, Gui C, et al. Level set evolution without re-initialization: a newvariational formulation[A]. Proceedings of IEEE Computer Society Conference onComputer Vision and Pattern Recognition[C]. IEEE,2005,1:430-436
[155]吴小庆.数学物理方程及其应用[M].北京:科学出版社,2008:14-19
[156] CASIA Gait Database[DB/OL]. http://www.sinobiometrics.com