基于运动补偿与RJ-MCMC结合的视频目标跟踪研究
|
摘要
视频中运动目标跟踪的研究是计算机视觉领域的一个重要课题,受到国内外学者的普遍关注。它在导弹视觉制导,机器人与无人机自主导航,军事目标的定位跟踪与识别,汽车自动驾驶,安全监控,智能交通,虚拟现实等方面有广阔的应用前景。在现实中,由于各种干扰的存在增加了视频目标检测和跟踪的难度。
视频中目标的检测是目标跟踪的基础,检测效果的不好则会给跟踪带来很大的影响。由于光照,遮挡,摄像机传感器带来的各种噪声使得目标检测变得很复杂,尤其是当摄像机运动时,背景和前景都随之运动,更不好区分前景和背景。本论文采取了一种基于运动补偿与自适应背景更新来检测动态场景中的运动目标。先计算前后两帧之间的金字塔光流场,光流场数据里既含有目标和背景的光流值也含有噪声的光流。由于事先不知道有多少个目标,也无法预测干扰所带来的噪声,因此就不能用类数目固定的k均值聚类。本论文采用了一种叫做leader-follow的在线聚类算法,该算法能根据输入数据自适应的进行分类,而无需事先知道要分成多少类。把背景的光流值的分离出来以后就得到了运动补偿值,运用此补偿值补偿上一帧图像,然后前一帧图像和当前帧图像的背景对齐了,这样就可以使用应用在静态场景中的背景更新方法来得到了动态场景的背景和前景了。
在目标跟踪领域,一般都从贝叶斯递归的角度来看目标跟踪问题,即根据一系列的观测结果来对某目标的状态进行估计。由于目标运动无规律难于用数学公式描述,而粒子滤波不需要事先对状态转移做任何假设,因此很多本领域的学者都把它拿来研究以用于目标跟踪。普通粒子滤波的粒子退化问题严重地限制了其基本方法的发展。MCMC作为粒子滤波器的一种它可以有效地解决上述的粒子退化问题正引起人们的关注。针对多目标的跟踪普通MCMC是无法应对的,因为当目标进出场景会引起解的状态变量空间变化,而RJ-MCMC则能有效应对此种情况。针对动态场景的多目标跟踪,本文提出了一种基于二次观测模型的RJ-MCMC粒子滤波方法第一次观测通过运动补偿值对运动模型实时修正使其逼近真实的运动方程,第二次观测即是RJ-MCMC粒子滤波步骤。时变的运动模型可以有效提高RJ-MCMC方法的效率,减少其无效的粒子点数,使其能更加快速地收敛到真实值。
为了对特征不明显或目标区域过小的目标进行跟踪,在系统的观测模型中本论文提出了基于颜色直方图与前景匹配结合的观测模型。此观测模型有效的应用了前景检测器中的前景信息。实验表明它可以有效地跟踪特征不明显或目标区域过小的目标。
Moving target tracking from videos is an important research topic of computer vision, which receives a wide attention from domestic and international scholars. It has broad application prospects in missile visual guidance, robot and autonomous navigation, military targets localization, tracking and identification, safety monitoring, intelligent driving, traffic, virtual reality, etc. In reality, the existences of various noises increase the difficulty of object detection and tracking from videos.
Target detection is the foundation of the target tracking, and the detection effect will bring significant influence for tracking. Various noises brought by light changing, target shade and camera sensors make target detection very complex, especially when the camera moves, background and foreground subsequently move accordingly in the images, so it brings more difficulties to distinguish the foreground from the background. This thesis adopts motion compensation and adaptive background updates to detect the target motion in dynamic scenes. Firstly calculate the pyramidal optical flow field of the two consecutive frames, the pyramidal optical flow field data contains not only target's and background's optical flow value but also that of noise. Because we cannot predict the number of targets and the interference caused by noise, we cannot use fixed number clustering algorithm. This thesis adopts an online clustering algorithm named leader-follow clustering algorithm, which can adaptively classify the pyramidal optical flow field data according to the input data. We get the compensation value when the optical flow value of the background is separated, then we use this value to compensate the previous frame. The background pixels of the previous frame are the same as the correspondence pixels in current frame after compensation. Then the background update algorithms applied in static scenes can be used to generate background.
In video target tracking field, usually target tracking can be treated as a Bayesian estimation, that is, estimate the states from the observation data. Because the erratic target motion is difficult to describe by mathematic formulae, and the particle filter do not need any prior assumptions to states transfer, so many scholars adopt it in target tracking field. The particle degradation problem in ordinary particle filter seriously limits its development. As a particle filter, MCMC can effectively solve this problem. But in multi-targets tracking, the number of targets in the scene is not fixed, moreover the MCMC can not deal with the dimensions of solution space that is alterable, so the MCMC can not deal with multi-target tracking. The RJ-MCMC is designed to solve this problem, so we adopt this method to deal with multi-target tracking. Aiming at the multi-target tracking in the dynamic scene, we proposed a new RJ-MCMC particle filtering method based on a twice-observation model, The first observation is to modify the motion model though motion compensation to approach the true motion equation of the targets, while the second observation is the RJ-MCMC particle filtering procedure. Time variant motion model can increase the efficiency of the RJ-MCMC algorithm, reduce the number of ineffective particles, and enable it convergence to the real value faster.
The features of targets may be obscure or its foreground regions are very small, this will increase the difficulty of tracking. Aiming to this problem we proposed a joint color histogram with foreground appearance comparability model. This observation model makes use of foreground information to enhance the features of targets. Experiments show that this improved observation model can effectively handle little targets' tracking.
视频中目标的检测是目标跟踪的基础,检测效果的不好则会给跟踪带来很大的影响。由于光照,遮挡,摄像机传感器带来的各种噪声使得目标检测变得很复杂,尤其是当摄像机运动时,背景和前景都随之运动,更不好区分前景和背景。本论文采取了一种基于运动补偿与自适应背景更新来检测动态场景中的运动目标。先计算前后两帧之间的金字塔光流场,光流场数据里既含有目标和背景的光流值也含有噪声的光流。由于事先不知道有多少个目标,也无法预测干扰所带来的噪声,因此就不能用类数目固定的k均值聚类。本论文采用了一种叫做leader-follow的在线聚类算法,该算法能根据输入数据自适应的进行分类,而无需事先知道要分成多少类。把背景的光流值的分离出来以后就得到了运动补偿值,运用此补偿值补偿上一帧图像,然后前一帧图像和当前帧图像的背景对齐了,这样就可以使用应用在静态场景中的背景更新方法来得到了动态场景的背景和前景了。
在目标跟踪领域,一般都从贝叶斯递归的角度来看目标跟踪问题,即根据一系列的观测结果来对某目标的状态进行估计。由于目标运动无规律难于用数学公式描述,而粒子滤波不需要事先对状态转移做任何假设,因此很多本领域的学者都把它拿来研究以用于目标跟踪。普通粒子滤波的粒子退化问题严重地限制了其基本方法的发展。MCMC作为粒子滤波器的一种它可以有效地解决上述的粒子退化问题正引起人们的关注。针对多目标的跟踪普通MCMC是无法应对的,因为当目标进出场景会引起解的状态变量空间变化,而RJ-MCMC则能有效应对此种情况。针对动态场景的多目标跟踪,本文提出了一种基于二次观测模型的RJ-MCMC粒子滤波方法第一次观测通过运动补偿值对运动模型实时修正使其逼近真实的运动方程,第二次观测即是RJ-MCMC粒子滤波步骤。时变的运动模型可以有效提高RJ-MCMC方法的效率,减少其无效的粒子点数,使其能更加快速地收敛到真实值。
为了对特征不明显或目标区域过小的目标进行跟踪,在系统的观测模型中本论文提出了基于颜色直方图与前景匹配结合的观测模型。此观测模型有效的应用了前景检测器中的前景信息。实验表明它可以有效地跟踪特征不明显或目标区域过小的目标。
Moving target tracking from videos is an important research topic of computer vision, which receives a wide attention from domestic and international scholars. It has broad application prospects in missile visual guidance, robot and autonomous navigation, military targets localization, tracking and identification, safety monitoring, intelligent driving, traffic, virtual reality, etc. In reality, the existences of various noises increase the difficulty of object detection and tracking from videos.
Target detection is the foundation of the target tracking, and the detection effect will bring significant influence for tracking. Various noises brought by light changing, target shade and camera sensors make target detection very complex, especially when the camera moves, background and foreground subsequently move accordingly in the images, so it brings more difficulties to distinguish the foreground from the background. This thesis adopts motion compensation and adaptive background updates to detect the target motion in dynamic scenes. Firstly calculate the pyramidal optical flow field of the two consecutive frames, the pyramidal optical flow field data contains not only target's and background's optical flow value but also that of noise. Because we cannot predict the number of targets and the interference caused by noise, we cannot use fixed number clustering algorithm. This thesis adopts an online clustering algorithm named leader-follow clustering algorithm, which can adaptively classify the pyramidal optical flow field data according to the input data. We get the compensation value when the optical flow value of the background is separated, then we use this value to compensate the previous frame. The background pixels of the previous frame are the same as the correspondence pixels in current frame after compensation. Then the background update algorithms applied in static scenes can be used to generate background.
In video target tracking field, usually target tracking can be treated as a Bayesian estimation, that is, estimate the states from the observation data. Because the erratic target motion is difficult to describe by mathematic formulae, and the particle filter do not need any prior assumptions to states transfer, so many scholars adopt it in target tracking field. The particle degradation problem in ordinary particle filter seriously limits its development. As a particle filter, MCMC can effectively solve this problem. But in multi-targets tracking, the number of targets in the scene is not fixed, moreover the MCMC can not deal with the dimensions of solution space that is alterable, so the MCMC can not deal with multi-target tracking. The RJ-MCMC is designed to solve this problem, so we adopt this method to deal with multi-target tracking. Aiming at the multi-target tracking in the dynamic scene, we proposed a new RJ-MCMC particle filtering method based on a twice-observation model, The first observation is to modify the motion model though motion compensation to approach the true motion equation of the targets, while the second observation is the RJ-MCMC particle filtering procedure. Time variant motion model can increase the efficiency of the RJ-MCMC algorithm, reduce the number of ineffective particles, and enable it convergence to the real value faster.
The features of targets may be obscure or its foreground regions are very small, this will increase the difficulty of tracking. Aiming to this problem we proposed a joint color histogram with foreground appearance comparability model. This observation model makes use of foreground information to enhance the features of targets. Experiments show that this improved observation model can effectively handle little targets' tracking.
引文
[1]Collins et al. A system for Video Surveillance and System Monitoring. VSAM Final Report, Technical report CMU-RI-TR-00-12 Carnegie University, May, 2000.
[2]张江山,朱光喜.一种基于Kalman滤波的视频对象跟踪方法[J].中国图像图形学报,2002,7(6)：606-609.
[3]TEKALP A. M. Digital processing[M]. Prentice Hall,1995.
[4]Marr D. Vision. San Francisco:Freeman W.H. and Company,1982.
[5]C. stauffer, W. grimson. Adaptive Background Mixture Models for Real-time Tracking[C]. In Proc IEEE Conference on Computer Vision and pattern Recognition, Fort Collins, Calorado,1999,2:246-252.
[6]Kalman R E. A new approach to linear filtering and prediction problems. Transactions of the ASME, Journal of Basic Engineering,1960.
[7]Barrow H G, Tenenbaum J. M. Computational vision. Proceedings of the IEEE, 1981.69(5):572-595.
[8]Steven M Kay. Fundamentals of Statistical Signal Processing [M]. Macmillan, New York,2002.
[9]Anderson B D O, Moore J B, Eslami, M. Optimal Filtering. IEEE Transactions on Systems, Man and Cybernetics,1982.12(2):235-236.
[10]Handschin J E. Monte Carlo techniques for prediction and filtering of non-linear stochastic processes[J]. Automatica,1970,6(3):555-563.
[11]Horn, Berthod, K.P.Schunck, Brian G. Determining optical flow[J]. Artificial Intelligence,1981,17(1-3):185-204.
[12]马颂德,张正友.计算机视觉—计算理论与算法基础[M].北京科学出版社,1998：235-239
[13]Stuart p.Lloyd. Least Squares Quantization In PCM. IEEE transaction on information theory, IT-2:129-137,1982.
[14]Leonard Kaufman and Peter J. Rousseeuw. Finding Group in Data:An Introduction to Cluster Analysis. [M] Wiley, new York,1990
[15]Djuric P. M, Kotecha J. H, Jianqui, Z, et al. Particle filtering[J]. Signal Processing Magazine, IEEE,2003.20(5):19-38.
[16]章毓晋 图象工程(下册).图像理解与计算机视觉[M].北京：清华大学出 版社,2000.
[15]陈乐,吕文阁,丁少华.角点检测技术研究进展[J]. 自动化技术与应用2005,24(5)：1-4.
[17]Bouguet J Y. Pyramidal Implementation of the Lucas Kanade Feature Tracker: Description of the algorithm[J]. Microprocessor Research Labs, Intel Corporation,1999.
[18]G. Adiv. Determining Three-Dimensional Motion and Structure from Optical Flow Generated by Several Moving Objects[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol.7, No.4,1985, p.384-401.
[19]贾云得.扔器视觉北京：科学出版社,2000.
[20]Haitao J, Mei X, Li R. An improved global motion estimation for practical objection detection. International Conference on Information and Automation, 2008:1159-1162.
[21]RICHARD O, DUDA, PETER E, et al. Pattern classifieation [M]. New York:John Wiley & Sons Inc.2001.
[22]WILLIAM H, DAY E, HERBERT E. Efficient algorithms for agglomerative hierarcal clustering methods [J]. Journal of classification,1984,1(1):7-24.
[23]李庆扬,王能超,易大义.数值分析[M]北京：清华大学出版社.2008.
[24]Tao Y, Stan Z L, Quan P, et al. Real-time and accurate segmentation of moving objects in dynamic scene[C]. Proceedings of the ACM 2nd international workshop on Video surveillance & sensor networks,2004.
[25]M. I. Ribeiro, Kalman and Extended Kalman Filters:Concept, Derivation and Properties [M]. Institute for Systems and Robotics Instituto Superior T'ecnico, Av. Rovisco Pais,1, February 2004.
[26]S.J. Julier and J.K. Uhlmann, A new extension of the Kalman filter to nonlinear systems[C]. Proceedings of Aerospace Sense:The 11th Int. Symp. On Aerospace/Defense Sensing, Simulation and Controls,1997.
[27]Li P, Zhang T, Pece A E. Visual contour tracking based on particle filters[J]. Image and Vision Computing,2003,21(1):111-123.
[28]Amlampalam M S, Maskell S, Gordon N, Clapp T. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking[J]. IEEE Transactions on Signal Processing,2002,50(2):174-188.
[29]Liu J, Chen.R. Sequential monte carlo methods for dynamic systems[J]. Journal of the American Statistical Association, September 1998,93, (443): 1032-1044.
[30]W.K. Hastings. Monte Carlo sampling methods using Markov chains and their applications. Biometrika,57:97-109 1970.
[31]W.R. Gilks, S. Richardson, and D.J. Spiegelhalter, editors. Markov chain Monte Carlo in practice[M]. Chapman and Hall,1996.
[32]C. Berzuini, N. G. Best, W.R. Gilks, and C. Larizza. Dynamic conditional independence models and Markov chain Monte Carlo methods[J]. Journal of the American Statistical Association,92:1403-1412 1996.
[33]P.J.Green. Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika,82:711-732 1995.
[34]P.J. Green. Trans-dimensional Markov chain Monte Carlo. Chapter in Highly Structured Stochastic Systems,2003.
[35]Rathi Y, Vaswani N, Tannenbaum A, et al. Tracking Deforming Objects Using Particle Filtering for Geometric Active Contours[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007.29(8):1470-1475.
[36]Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003.25(5): 564-577.
[37]冈萨雷斯.数字图像处理[M]北京：电子工业出版社,2004.
[38]http://en. wikipedia. org/wiki/Bhattacharyya_coefficient.
[39]Kailath, T. The Divergence and Bhattacharyya Distance Measures in Signal Selection[J]. IEEE Transactions on Communication Technology 15 (1):52-60. 1967.
[40]Zia Khan, T. Balch, F. Dellaert. MCMC-Based Particle Filtering for Tracking a Variable Number of Interacting Targets[C]. In Proceedings of the European Conference on Computer Vision (ECCV), Prague, May 2004.
[41]张苗辉,刘先省.基于MCMC无味粒子滤波的目标跟踪算法.[J]系统工程与电子技术,2009,8(8) vol.31no.8.
[42]T. Zhao and R. Nevatia. Tracking multiple humans in crowded environment[C]. In IEEE Conf on Computer Vision and Pattern Recognition (CVPR),2004..
[43]Zia Khan, T. Balch, F. Dellaert. MCMC-Based Particle Filtering for Tracking a Variable Number of Interacting Targets[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27:1805-1819.
[44]胡士强,敬忠良.粒子滤波算法综述[J]。控制与决策,2005,20(4)：361-365.
[45]Metropolis N, Rosenbluth A W, Rosenbluth M N, et al. Equation of State Calculations by Fast Computing Machines[J].The Journal of Chemical Physics, 1953.21(6):1087-1092.
[46]Gordon N. J, Salmond D. J, Smith A. F. M. Novel approach to nonlinear/non-Gaussian Bayesian state estimation[C]. Radar and Signal Processing, IEE Proceedings,1993.140(2):107-113.
[47]MacCormick J, Blake A. A Probabilistic Exclusion Principle for Tracking Multiple Objects[C]. Int. J. Computer Vision,2000.39(1):57-71.
[48]Djuric P. M, Kotecha J. H, Jianqui, Z, et al. Particle filtering[J]. Signal Processing Magazine, IEEE,2003.20(5):19-38.
[49]Doucet A, Godsill S, Andrieu C. On sequential Monte Carlo sampling methods for Bayesian filtering. Statistics and Computing,2000.10(3):197-208.
[50]Pengwei Liu, Huiyuan Wang, Xiaojuan Wu, Wei Qiao. Motion Compensation Based Detecting and Tracking Targets in Dynamic Scene[C]. Proceedings of ICMTMA2010, IEEE Computer Society,2010,2:703-706.
[2]张江山,朱光喜.一种基于Kalman滤波的视频对象跟踪方法[J].中国图像图形学报,2002,7(6)：606-609.
[3]TEKALP A. M. Digital processing[M]. Prentice Hall,1995.
[4]Marr D. Vision. San Francisco:Freeman W.H. and Company,1982.
[5]C. stauffer, W. grimson. Adaptive Background Mixture Models for Real-time Tracking[C]. In Proc IEEE Conference on Computer Vision and pattern Recognition, Fort Collins, Calorado,1999,2:246-252.
[6]Kalman R E. A new approach to linear filtering and prediction problems. Transactions of the ASME, Journal of Basic Engineering,1960.
[7]Barrow H G, Tenenbaum J. M. Computational vision. Proceedings of the IEEE, 1981.69(5):572-595.
[8]Steven M Kay. Fundamentals of Statistical Signal Processing [M]. Macmillan, New York,2002.
[9]Anderson B D O, Moore J B, Eslami, M. Optimal Filtering. IEEE Transactions on Systems, Man and Cybernetics,1982.12(2):235-236.
[10]Handschin J E. Monte Carlo techniques for prediction and filtering of non-linear stochastic processes[J]. Automatica,1970,6(3):555-563.
[11]Horn, Berthod, K.P.Schunck, Brian G. Determining optical flow[J]. Artificial Intelligence,1981,17(1-3):185-204.
[12]马颂德,张正友.计算机视觉—计算理论与算法基础[M].北京科学出版社,1998：235-239
[13]Stuart p.Lloyd. Least Squares Quantization In PCM. IEEE transaction on information theory, IT-2:129-137,1982.
[14]Leonard Kaufman and Peter J. Rousseeuw. Finding Group in Data:An Introduction to Cluster Analysis. [M] Wiley, new York,1990
[15]Djuric P. M, Kotecha J. H, Jianqui, Z, et al. Particle filtering[J]. Signal Processing Magazine, IEEE,2003.20(5):19-38.
[16]章毓晋 图象工程(下册).图像理解与计算机视觉[M].北京：清华大学出 版社,2000.
[15]陈乐,吕文阁,丁少华.角点检测技术研究进展[J]. 自动化技术与应用2005,24(5)：1-4.
[17]Bouguet J Y. Pyramidal Implementation of the Lucas Kanade Feature Tracker: Description of the algorithm[J]. Microprocessor Research Labs, Intel Corporation,1999.
[18]G. Adiv. Determining Three-Dimensional Motion and Structure from Optical Flow Generated by Several Moving Objects[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, Vol.7, No.4,1985, p.384-401.
[19]贾云得.扔器视觉北京：科学出版社,2000.
[20]Haitao J, Mei X, Li R. An improved global motion estimation for practical objection detection. International Conference on Information and Automation, 2008:1159-1162.
[21]RICHARD O, DUDA, PETER E, et al. Pattern classifieation [M]. New York:John Wiley & Sons Inc.2001.
[22]WILLIAM H, DAY E, HERBERT E. Efficient algorithms for agglomerative hierarcal clustering methods [J]. Journal of classification,1984,1(1):7-24.
[23]李庆扬,王能超,易大义.数值分析[M]北京：清华大学出版社.2008.
[24]Tao Y, Stan Z L, Quan P, et al. Real-time and accurate segmentation of moving objects in dynamic scene[C]. Proceedings of the ACM 2nd international workshop on Video surveillance & sensor networks,2004.
[25]M. I. Ribeiro, Kalman and Extended Kalman Filters:Concept, Derivation and Properties [M]. Institute for Systems and Robotics Instituto Superior T'ecnico, Av. Rovisco Pais,1, February 2004.
[26]S.J. Julier and J.K. Uhlmann, A new extension of the Kalman filter to nonlinear systems[C]. Proceedings of Aerospace Sense:The 11th Int. Symp. On Aerospace/Defense Sensing, Simulation and Controls,1997.
[27]Li P, Zhang T, Pece A E. Visual contour tracking based on particle filters[J]. Image and Vision Computing,2003,21(1):111-123.
[28]Amlampalam M S, Maskell S, Gordon N, Clapp T. A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking[J]. IEEE Transactions on Signal Processing,2002,50(2):174-188.
[29]Liu J, Chen.R. Sequential monte carlo methods for dynamic systems[J]. Journal of the American Statistical Association, September 1998,93, (443): 1032-1044.
[30]W.K. Hastings. Monte Carlo sampling methods using Markov chains and their applications. Biometrika,57:97-109 1970.
[31]W.R. Gilks, S. Richardson, and D.J. Spiegelhalter, editors. Markov chain Monte Carlo in practice[M]. Chapman and Hall,1996.
[32]C. Berzuini, N. G. Best, W.R. Gilks, and C. Larizza. Dynamic conditional independence models and Markov chain Monte Carlo methods[J]. Journal of the American Statistical Association,92:1403-1412 1996.
[33]P.J.Green. Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika,82:711-732 1995.
[34]P.J. Green. Trans-dimensional Markov chain Monte Carlo. Chapter in Highly Structured Stochastic Systems,2003.
[35]Rathi Y, Vaswani N, Tannenbaum A, et al. Tracking Deforming Objects Using Particle Filtering for Geometric Active Contours[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007.29(8):1470-1475.
[36]Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003.25(5): 564-577.
[37]冈萨雷斯.数字图像处理[M]北京：电子工业出版社,2004.
[38]http://en. wikipedia. org/wiki/Bhattacharyya_coefficient.
[39]Kailath, T. The Divergence and Bhattacharyya Distance Measures in Signal Selection[J]. IEEE Transactions on Communication Technology 15 (1):52-60. 1967.
[40]Zia Khan, T. Balch, F. Dellaert. MCMC-Based Particle Filtering for Tracking a Variable Number of Interacting Targets[C]. In Proceedings of the European Conference on Computer Vision (ECCV), Prague, May 2004.
[41]张苗辉,刘先省.基于MCMC无味粒子滤波的目标跟踪算法.[J]系统工程与电子技术,2009,8(8) vol.31no.8.
[42]T. Zhao and R. Nevatia. Tracking multiple humans in crowded environment[C]. In IEEE Conf on Computer Vision and Pattern Recognition (CVPR),2004..
[43]Zia Khan, T. Balch, F. Dellaert. MCMC-Based Particle Filtering for Tracking a Variable Number of Interacting Targets[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27:1805-1819.
[44]胡士强,敬忠良.粒子滤波算法综述[J]。控制与决策,2005,20(4)：361-365.
[45]Metropolis N, Rosenbluth A W, Rosenbluth M N, et al. Equation of State Calculations by Fast Computing Machines[J].The Journal of Chemical Physics, 1953.21(6):1087-1092.
[46]Gordon N. J, Salmond D. J, Smith A. F. M. Novel approach to nonlinear/non-Gaussian Bayesian state estimation[C]. Radar and Signal Processing, IEE Proceedings,1993.140(2):107-113.
[47]MacCormick J, Blake A. A Probabilistic Exclusion Principle for Tracking Multiple Objects[C]. Int. J. Computer Vision,2000.39(1):57-71.
[48]Djuric P. M, Kotecha J. H, Jianqui, Z, et al. Particle filtering[J]. Signal Processing Magazine, IEEE,2003.20(5):19-38.
[49]Doucet A, Godsill S, Andrieu C. On sequential Monte Carlo sampling methods for Bayesian filtering. Statistics and Computing,2000.10(3):197-208.
[50]Pengwei Liu, Huiyuan Wang, Xiaojuan Wu, Wei Qiao. Motion Compensation Based Detecting and Tracking Targets in Dynamic Scene[C]. Proceedings of ICMTMA2010, IEEE Computer Society,2010,2:703-706.