基于改进粒子滤波器目标跟踪算法研究
|
摘要
目标跟踪被广泛应用于视频监控、安防系统、智能交通系统及机器人技术等领域,是一些需要确定目标位置、运动和身份等智能系统的核心组成部分,可以说是一个非常棘手而富有挑战性的课题。由于实际应用中存在诸如摄像头运动、目标不稳定、背景复杂以及其他相似移动物体等的困难的情况,人们很难找到一个广泛适用的鲁棒性高的跟踪算法。四十多年前,卡尔曼先生提出了卡尔曼滤波算法,它简单而便于实现,是解决线性高斯环境下的问题的最佳方法。近年来由于技术的发展和应用的需要,出现了一个研究非线性非高斯环境下滤波算法的高潮。
本文首先介绍了近年来常见的一些目标跟踪滤波算法——卡尔曼滤波器(Kalman filter, KF)、扩展卡尔曼滤波器(Extended Kalman filter, EKF),无敏卡尔曼滤波器(Unscented Kalman filter, UKF),粒子滤波器(Particle filter, PF)。KF简单而优雅,是线性高斯环境下的最佳递归贝叶斯滤波器。EKF利用泰勒级数方法,将非线性问题转化到线性空间,再利用卡尔曼滤波器进行估计滤波,并达到一阶估计精度。UKF通过固定样本集达到对状态概率分布的近似,在精度和计算量上较之EKF优秀,但它是利用高斯分布来逼近系统状态的后验概率密度,在复杂的环境中表现差。PF是一种采用蒙特卡罗采样的贝叶斯滤波方法,它将复杂的目标状态分布表示为一组加权值(称为粒子),通过寻找在粒子滤波分布中最大权重的粒子来确定目标最可能所处的状态分布,已成为复杂环境下进行目标跟踪的最好的方法。本论文通过量测非线性模型(正切)的对比实验,证明了PF在非线性环境下有着最优异表现,UKF表现较之EKF优异,而EKF优于KF,与理论分析的结果一致。
在目标跟踪系统中选取描述目标的特征是一个棘手的问题,使用更多的描述目标的特征可以有效提高跟踪的准确性,但会增加计算机的计算量和计算时间,只能取实时性和准确性的折衷。目标颜色直方图特征具有稳定性高、计算量较小的特点,已成为主流的描述目标的特征。描述目标颜色特征的颜色空间有很多种,本文介绍了常见的RGB空间、CMYK空间、HSV空间。其中HSV空间更符合人眼感知色彩的方式,此空间模型具有线性伸缩性良好,色差与颜色分量在相应值上的欧几里德距离成比例等优点。但单一的颜色直方图特征对背景光照变化敏感,而且当有相似颜色干扰信息时,跟踪的准确性大大降低。而目标的结构性特征主要有矩特征,矩特征具有平移、旋转、尺度等不变特性,被广泛应用在图像匹配、姿态识别等领域。
本文结合粒子滤波算法提出一种基于融合目标不变矩特征和颜色直方图特征的目标跟踪方法,其中目标颜色直方图特征的计算是在HSV空间进行的。该方法对目标的颜色信息和结构信息进行融合并建立目标模型,粒子权值的融合比例由应用环境决定,在系统更新过程中比较目标与粒子间的欧几里德距离,淘汰劣质的粒子,增加了粒子的可靠度,减少了噪声的影响。实验数据表明此改进方法有效克服了单一颜色特征模型在应用中的不足,并能够在不影响实时性的基础上提高跟踪的有效性。
Target tracking is the core components of intelligent systems to determine location, movement and identity targets, which is widely used in the field of video surveillance, security systems and intelligent transportation system. It is really a tough job to realize this system and find a widely used and high robust tracking algorithm due to camera movement, target instability, complexity of background and moving similarity. It is hard to find. Kalman filter algorithm, proposed by Mr. Kalman more than 40 years ago, is the best way to solve the problem in the linear Gaussian environment. However, in order to meet technology and application needs, there has been an emergence of studying the nonlinear non-Gaussian filtering algorithm recent years.
First, this thesis introduced the common target tracking filter algorithm proposed recently, such as Kalman Filter (KF), extended Kalman filter (EKF), Unscented Filter (UKF) and particle filter (PF) algorithm. Simple and elegant, KF is the best recursive Bayesian estimator in linear Gaussian environment. By using Taylor series, EKF transforms nonlinear problem into linear space, then using Kalman filter to estimate the results to achieve the first order accuracy. Through the fixed sample set to approximate the probability distribution of the state, UKF is better than the EKF on precision and quantity. Nonetheless, as using Gaussian posterior probability density to approximate the system state, it is poor to perform in complex environment. Particle filter (PF) is a Bayesian filtering adopted by Monte Carlo sampling method. The complex target state distribution is expressed as a set of weights (called particle) in this filter. By finding the largest weight particles in the particle filter to determine the most likely target has been proved as the best way to track target in a complex environment. By measuring the nonlinear model (tangent), this thesis demonstrated that the particle filter has the most outstanding performance dealing with the nonlinear situations, and UKF has superior performance than EKF, EKF is better than KF, which is identical to the theoretical analysis.
Second, it is a tough job to select the characteristics of targets in target tracking system. If targets have more features, tracking accuracy could be effectively improved, however, computing quantity and calculation time would also increasing. It is imperative for us to take compromise of real-time and accuracy. As high stability and low computational characteristics, color histogram feature are becoming a main feature to describe targets. This thesis introduced RGB space, CMYK space, HSV space. All these spaces, HSV space is more suitable for human to perceive color. Also, this model has good linear scalability advantages. However, single color histogram is sensitive to the background illumination and what's more, tracking accuracy could reduced significantly when this system is interfered with similar color objects. By contract, as objects moments feature has structural characteristics, which have the properties of translation, rotation and scale invariant, etc. It is widely used in image matching and gesture recognition.
Finally, by combining both properties of target color and moment invariant this thesis presented an improved particle filter based on method, the characteristic of the color histogram is carried out in HSV space. The weights of the particle are determined by application environment. Further, by determined Euclidean distance properties in the process of replacement, poor quality particles were washed out, reliability of particles were increased and the impact of noises were reduced. The experimental results have been demonstrated that this method ameliorated the interference immunity of the single color property for tracking target. In addition, this method also improved the tracking accuracy and robustness while not affecting the real-time characteristics.
本文首先介绍了近年来常见的一些目标跟踪滤波算法——卡尔曼滤波器(Kalman filter, KF)、扩展卡尔曼滤波器(Extended Kalman filter, EKF),无敏卡尔曼滤波器(Unscented Kalman filter, UKF),粒子滤波器(Particle filter, PF)。KF简单而优雅,是线性高斯环境下的最佳递归贝叶斯滤波器。EKF利用泰勒级数方法,将非线性问题转化到线性空间,再利用卡尔曼滤波器进行估计滤波,并达到一阶估计精度。UKF通过固定样本集达到对状态概率分布的近似,在精度和计算量上较之EKF优秀,但它是利用高斯分布来逼近系统状态的后验概率密度,在复杂的环境中表现差。PF是一种采用蒙特卡罗采样的贝叶斯滤波方法,它将复杂的目标状态分布表示为一组加权值(称为粒子),通过寻找在粒子滤波分布中最大权重的粒子来确定目标最可能所处的状态分布,已成为复杂环境下进行目标跟踪的最好的方法。本论文通过量测非线性模型(正切)的对比实验,证明了PF在非线性环境下有着最优异表现,UKF表现较之EKF优异,而EKF优于KF,与理论分析的结果一致。
在目标跟踪系统中选取描述目标的特征是一个棘手的问题,使用更多的描述目标的特征可以有效提高跟踪的准确性,但会增加计算机的计算量和计算时间,只能取实时性和准确性的折衷。目标颜色直方图特征具有稳定性高、计算量较小的特点,已成为主流的描述目标的特征。描述目标颜色特征的颜色空间有很多种,本文介绍了常见的RGB空间、CMYK空间、HSV空间。其中HSV空间更符合人眼感知色彩的方式,此空间模型具有线性伸缩性良好,色差与颜色分量在相应值上的欧几里德距离成比例等优点。但单一的颜色直方图特征对背景光照变化敏感,而且当有相似颜色干扰信息时,跟踪的准确性大大降低。而目标的结构性特征主要有矩特征,矩特征具有平移、旋转、尺度等不变特性,被广泛应用在图像匹配、姿态识别等领域。
本文结合粒子滤波算法提出一种基于融合目标不变矩特征和颜色直方图特征的目标跟踪方法,其中目标颜色直方图特征的计算是在HSV空间进行的。该方法对目标的颜色信息和结构信息进行融合并建立目标模型,粒子权值的融合比例由应用环境决定,在系统更新过程中比较目标与粒子间的欧几里德距离,淘汰劣质的粒子,增加了粒子的可靠度,减少了噪声的影响。实验数据表明此改进方法有效克服了单一颜色特征模型在应用中的不足,并能够在不影响实时性的基础上提高跟踪的有效性。
Target tracking is the core components of intelligent systems to determine location, movement and identity targets, which is widely used in the field of video surveillance, security systems and intelligent transportation system. It is really a tough job to realize this system and find a widely used and high robust tracking algorithm due to camera movement, target instability, complexity of background and moving similarity. It is hard to find. Kalman filter algorithm, proposed by Mr. Kalman more than 40 years ago, is the best way to solve the problem in the linear Gaussian environment. However, in order to meet technology and application needs, there has been an emergence of studying the nonlinear non-Gaussian filtering algorithm recent years.
First, this thesis introduced the common target tracking filter algorithm proposed recently, such as Kalman Filter (KF), extended Kalman filter (EKF), Unscented Filter (UKF) and particle filter (PF) algorithm. Simple and elegant, KF is the best recursive Bayesian estimator in linear Gaussian environment. By using Taylor series, EKF transforms nonlinear problem into linear space, then using Kalman filter to estimate the results to achieve the first order accuracy. Through the fixed sample set to approximate the probability distribution of the state, UKF is better than the EKF on precision and quantity. Nonetheless, as using Gaussian posterior probability density to approximate the system state, it is poor to perform in complex environment. Particle filter (PF) is a Bayesian filtering adopted by Monte Carlo sampling method. The complex target state distribution is expressed as a set of weights (called particle) in this filter. By finding the largest weight particles in the particle filter to determine the most likely target has been proved as the best way to track target in a complex environment. By measuring the nonlinear model (tangent), this thesis demonstrated that the particle filter has the most outstanding performance dealing with the nonlinear situations, and UKF has superior performance than EKF, EKF is better than KF, which is identical to the theoretical analysis.
Second, it is a tough job to select the characteristics of targets in target tracking system. If targets have more features, tracking accuracy could be effectively improved, however, computing quantity and calculation time would also increasing. It is imperative for us to take compromise of real-time and accuracy. As high stability and low computational characteristics, color histogram feature are becoming a main feature to describe targets. This thesis introduced RGB space, CMYK space, HSV space. All these spaces, HSV space is more suitable for human to perceive color. Also, this model has good linear scalability advantages. However, single color histogram is sensitive to the background illumination and what's more, tracking accuracy could reduced significantly when this system is interfered with similar color objects. By contract, as objects moments feature has structural characteristics, which have the properties of translation, rotation and scale invariant, etc. It is widely used in image matching and gesture recognition.
Finally, by combining both properties of target color and moment invariant this thesis presented an improved particle filter based on method, the characteristic of the color histogram is carried out in HSV space. The weights of the particle are determined by application environment. Further, by determined Euclidean distance properties in the process of replacement, poor quality particles were washed out, reliability of particles were increased and the impact of noises were reduced. The experimental results have been demonstrated that this method ameliorated the interference immunity of the single color property for tracking target. In addition, this method also improved the tracking accuracy and robustness while not affecting the real-time characteristics.
引文
[1]章毓晋.图像工程(上册)---图像处理和分析[M].北京：清华大学出版社,1999.
[2]A.Yilmaz, O.Javed, M.Shah. Object tracking:a survey[J]. ACM Computing Surveys,2006,38(4):1-45.
[3]D.Gavrila, V.Philomin. Real-time object detection for Smart Vehicles[J]. In Proceedings.of IEEE International Conference on Computer Vision,1999, 1:87-93.
[4]M.B.Van Leeuwen, F.C.A.Groen, A.Dev. Requirements for motion estimation in Image Sequences for Traffic Applications[J]. In Proceedings Of IEEE Instrumentation and Measurement Technology Conference,1999,145-150.
[5]盛骤,谢式千,潘承毅.概率论与数理统计[M].北京：高等教育出版,1989.
[6]Y.Bar-Shalom. Multitarget-multisensor-tracking:Applications and Advances Volume Ⅲ[M]. Boston:Artech House,2000:103-119.
[7]唐勇,张桂林,李利荣.基于光流的运动目标检测方法研究[C].全国冶金自动化信息网年会论文集,2004：987-990.
[8]周露平,陈会勇,方伟,等.基于Kalman滤波特征跟踪系统仿真技术及其应用[C].2005系统仿真技术及其应用学术交流会论文选编(第7卷),2005：263-267.
[9]李鹏飞,黄建军,黄敬雄,等.一种模糊Kalman滤波机动目标跟踪的新算法[J].数据采集与处理,2009,24(4)：483-486.
[10]庄严,站洪斌,王伟.基于加权颜色直方图和粒子滤波的彩色物体跟踪[J].决策与控制,2006,21(8)：868-872.
[11]解云霄,周兰江,王伟.基于主目标区域形状不变矩的图像匹配算法[J].人工智能及识别技术,2009,5(19)：5250-5270.
[12]邓自立.最优估计理论及其应用——建模、滤波、信息融合估计[M].哈尔滨：哈尔滨工业大学出版社,2005.
[13]A.Sluzek. Identification and inspection of 2-d objects using new moment-based shape descriptors[J]. Pattern Recognition Letters,1995,16(7):687-697.
[14]韩晓微.彩色图像处理关键技术研究[D]沈阳：东北大学,2004.
[15]N.Katja, K.M.Esther, V.G.Luc. Color features for tracking non-rigid objects[J]. Acta Automatica Sinica,2003,29(3):345-355.
[16]付永会,张风超,张宪民.一种改进的基于颜色直方图的实时目标跟踪算法[J].数据采集与处理,2001,16(3)：309-314.
[17]J.Schubert, H.Sidenbladh. Sequential clustering with particle filters estimating the number of clusters from data[C].In Proceedings of the 8th International Conference on Information Fusion volume 1, Philadelphia:International Society of Information Fusion,2005:122-129.
[18]P.Minvielle, A.D.Marrs, S.Maskell, et al. Joint target tracking and identification-part i:Sequential monte-carlo model-based approaches[C]. In Proceedings of the 8th International Conference on Information Fusion volume 1, Philadelphia:International Society of Information Fusion,2005:259-266.
[19]裴鹿成,张孝泽.蒙特卡罗方法及其在粒子输运问题中的应用[M].北京：北京科学出版社,1980.
[20]裴鹿成,王仲奇.蒙特卡罗方法及其应用[M].北京：北京海洋出版社,1998.
[21]J.S.Liu, R.Chen. Sequential Monte Carlo methods for dynamical systems[J]. Journal of the Naerican Statistical Association,1998,93:1032-1044.
[22]张丽静,孙杰,殷晓宇.基于HSV颜色空间的车牌定位方法[J].测控自动化,2008,24(3)：247-248.
[23]E.A.Wan, R.V.D.Merwe. The unscented Kalman filter for nonlinear estimation[C]. In Proceeding of IEEE Symposium on Adaptive Systems for Signal Processing, Communications, and Control Symposium. Lake Louise,2000:153-158.
[24]万莉,刘焰春,皮亦鸣.EKF、UKF、PF目标跟踪性能的比较[J].雷达科学与技术,2007,5(1)：13-16.
[25]W.J.Szajnowski, P.A.Ratliff. Implicit averaging and Delay Determination of Random Binary Waveforms[J]. IEEE Signal Processing Letters,2002, 9(7):193-195.
[26]S.J.Julier, J.K.Uhlmann. A new extension of the Kalman filtert on onlinear systems [C]. In Proceedings of the 11th Annual International Symposium on Aerospace/Defense Sensing, Simulation, and Controls, Orlando,1997:182-193.
[27]S.J.Julier, J.K.Uhlmann, H.F.Durrant-Whyte. A new method for the nonlinear transformation of means and covariances in filters and estimators[J]. IEEE Transactions on automatic control,2000,43(3):477-482.
[28]M.K.Hu. Visual patten recognition by moment invariants[J]. IEEE Transactions on information theory,1962, (8):179-187.
[29]何友,王国宏,陆大琻,等.多传感器信息融合及应用[M].北京：电子工业出版社,2000.
[30]王志贤.最优状态估计与系统辨识[M].西安：西北工业大学出版社,2004.
[31]Y.Rosenberg, M.Werman. Real-Time object tracking from a moving video camera:A software approach on a PC[C]. IEEE Workshop on applications of computer vision, princeton,1998:238-239.
[32]冈萨雷斯,伍兹.数字图像处理(第二版)[M]．北京：电子工业出版社,2007.
[33]N.J.Gordon, D.J.Salmond, A.F.M.Smith. Novel approach to nonlinear/ non-gaussian bayesian state estimation[J]. In IEEE Proceedings on Radar and Signal Processing,1993,140:107-113.
[34]G.Kitagawa. Montecarlo filter and smoother for non-gaussian nonlinear state space models[J]. Journal of Computational and Graphical Statistics,1996, 5(1):1-25.
[35]M.Isard, A.Blake. Condensation-conditional density propagation for visual tracking[J]. International Journal of Computer Vision,1998,29(1):5-28.
[36]朱民雄,宋屹晏.实时数据处理方法在目标跟踪中的应用[J].微机发展,2000,3：60-63.
[37]C.Musso, N.Oudjane, F.LeGland. Improving regularised particle filters[C] In Sequential MonteCarlo in Practice, New York:Springer-Verlag,2001.
[38]M.S.Arulamalam, S.Maskel, N.Gordon, et al. A tutorial on particle filter for online nolinear/non-gaussian bayesian tracking[J].IEEE Transactions on Signal Proceeding,2002,50(2):174-188.
[39]T. Kalinke, C.Tzomakas, W.V.Seelen. A texturebased object detection and an adaptive model-based classification[J]. In Proceeding On IEEE Intelligent Vehicles Symposium,1998,143-148.
[40]H.A.P.Blom, E.A.Bloem. Joint immpda particle filter[C]. In Proceeding 6th International Conference on information Fusion volume 1, Cairns Queensland, 2003:785-792.
[41]J.Maccormick. Stochastic Algorithms for visual tracking[M]. New York:Springer-Verlag,2002.
[42]R.Morales-Men'endez, N.de Freitas, D.Poole. Real-time monitoring of complex industrial processes with particle filters[C]. In Proceeding of the Conference on Neural Information Processing Systems (NIPS), Vancover,2002:1433-1440.
[43]胡明昊,任明武,杨静宇．一种基于直方图模式的运动目标实时跟踪算法[J].计算机工程与应用,2004,3：71-74.
[44]O.Williams, A.Blake, R.Cipolla. A sparse probabilistic learning algorithm for real-time tracking[C]. In Proceedings of the Ninth IEEE International Conference on Computer Vision, Nice:IEEE Computer Society,2003:353-361.
[45]K.Toyama, A.Blake. Probabilistic tracking with exemplars in a metric space[J]. International Journal of Computer Vison,2002,48(1):9-19.
[2]A.Yilmaz, O.Javed, M.Shah. Object tracking:a survey[J]. ACM Computing Surveys,2006,38(4):1-45.
[3]D.Gavrila, V.Philomin. Real-time object detection for Smart Vehicles[J]. In Proceedings.of IEEE International Conference on Computer Vision,1999, 1:87-93.
[4]M.B.Van Leeuwen, F.C.A.Groen, A.Dev. Requirements for motion estimation in Image Sequences for Traffic Applications[J]. In Proceedings Of IEEE Instrumentation and Measurement Technology Conference,1999,145-150.
[5]盛骤,谢式千,潘承毅.概率论与数理统计[M].北京：高等教育出版,1989.
[6]Y.Bar-Shalom. Multitarget-multisensor-tracking:Applications and Advances Volume Ⅲ[M]. Boston:Artech House,2000:103-119.
[7]唐勇,张桂林,李利荣.基于光流的运动目标检测方法研究[C].全国冶金自动化信息网年会论文集,2004：987-990.
[8]周露平,陈会勇,方伟,等.基于Kalman滤波特征跟踪系统仿真技术及其应用[C].2005系统仿真技术及其应用学术交流会论文选编(第7卷),2005：263-267.
[9]李鹏飞,黄建军,黄敬雄,等.一种模糊Kalman滤波机动目标跟踪的新算法[J].数据采集与处理,2009,24(4)：483-486.
[10]庄严,站洪斌,王伟.基于加权颜色直方图和粒子滤波的彩色物体跟踪[J].决策与控制,2006,21(8)：868-872.
[11]解云霄,周兰江,王伟.基于主目标区域形状不变矩的图像匹配算法[J].人工智能及识别技术,2009,5(19)：5250-5270.
[12]邓自立.最优估计理论及其应用——建模、滤波、信息融合估计[M].哈尔滨：哈尔滨工业大学出版社,2005.
[13]A.Sluzek. Identification and inspection of 2-d objects using new moment-based shape descriptors[J]. Pattern Recognition Letters,1995,16(7):687-697.
[14]韩晓微.彩色图像处理关键技术研究[D]沈阳：东北大学,2004.
[15]N.Katja, K.M.Esther, V.G.Luc. Color features for tracking non-rigid objects[J]. Acta Automatica Sinica,2003,29(3):345-355.
[16]付永会,张风超,张宪民.一种改进的基于颜色直方图的实时目标跟踪算法[J].数据采集与处理,2001,16(3)：309-314.
[17]J.Schubert, H.Sidenbladh. Sequential clustering with particle filters estimating the number of clusters from data[C].In Proceedings of the 8th International Conference on Information Fusion volume 1, Philadelphia:International Society of Information Fusion,2005:122-129.
[18]P.Minvielle, A.D.Marrs, S.Maskell, et al. Joint target tracking and identification-part i:Sequential monte-carlo model-based approaches[C]. In Proceedings of the 8th International Conference on Information Fusion volume 1, Philadelphia:International Society of Information Fusion,2005:259-266.
[19]裴鹿成,张孝泽.蒙特卡罗方法及其在粒子输运问题中的应用[M].北京：北京科学出版社,1980.
[20]裴鹿成,王仲奇.蒙特卡罗方法及其应用[M].北京：北京海洋出版社,1998.
[21]J.S.Liu, R.Chen. Sequential Monte Carlo methods for dynamical systems[J]. Journal of the Naerican Statistical Association,1998,93:1032-1044.
[22]张丽静,孙杰,殷晓宇.基于HSV颜色空间的车牌定位方法[J].测控自动化,2008,24(3)：247-248.
[23]E.A.Wan, R.V.D.Merwe. The unscented Kalman filter for nonlinear estimation[C]. In Proceeding of IEEE Symposium on Adaptive Systems for Signal Processing, Communications, and Control Symposium. Lake Louise,2000:153-158.
[24]万莉,刘焰春,皮亦鸣.EKF、UKF、PF目标跟踪性能的比较[J].雷达科学与技术,2007,5(1)：13-16.
[25]W.J.Szajnowski, P.A.Ratliff. Implicit averaging and Delay Determination of Random Binary Waveforms[J]. IEEE Signal Processing Letters,2002, 9(7):193-195.
[26]S.J.Julier, J.K.Uhlmann. A new extension of the Kalman filtert on onlinear systems [C]. In Proceedings of the 11th Annual International Symposium on Aerospace/Defense Sensing, Simulation, and Controls, Orlando,1997:182-193.
[27]S.J.Julier, J.K.Uhlmann, H.F.Durrant-Whyte. A new method for the nonlinear transformation of means and covariances in filters and estimators[J]. IEEE Transactions on automatic control,2000,43(3):477-482.
[28]M.K.Hu. Visual patten recognition by moment invariants[J]. IEEE Transactions on information theory,1962, (8):179-187.
[29]何友,王国宏,陆大琻,等.多传感器信息融合及应用[M].北京：电子工业出版社,2000.
[30]王志贤.最优状态估计与系统辨识[M].西安：西北工业大学出版社,2004.
[31]Y.Rosenberg, M.Werman. Real-Time object tracking from a moving video camera:A software approach on a PC[C]. IEEE Workshop on applications of computer vision, princeton,1998:238-239.
[32]冈萨雷斯,伍兹.数字图像处理(第二版)[M]．北京：电子工业出版社,2007.
[33]N.J.Gordon, D.J.Salmond, A.F.M.Smith. Novel approach to nonlinear/ non-gaussian bayesian state estimation[J]. In IEEE Proceedings on Radar and Signal Processing,1993,140:107-113.
[34]G.Kitagawa. Montecarlo filter and smoother for non-gaussian nonlinear state space models[J]. Journal of Computational and Graphical Statistics,1996, 5(1):1-25.
[35]M.Isard, A.Blake. Condensation-conditional density propagation for visual tracking[J]. International Journal of Computer Vision,1998,29(1):5-28.
[36]朱民雄,宋屹晏.实时数据处理方法在目标跟踪中的应用[J].微机发展,2000,3：60-63.
[37]C.Musso, N.Oudjane, F.LeGland. Improving regularised particle filters[C] In Sequential MonteCarlo in Practice, New York:Springer-Verlag,2001.
[38]M.S.Arulamalam, S.Maskel, N.Gordon, et al. A tutorial on particle filter for online nolinear/non-gaussian bayesian tracking[J].IEEE Transactions on Signal Proceeding,2002,50(2):174-188.
[39]T. Kalinke, C.Tzomakas, W.V.Seelen. A texturebased object detection and an adaptive model-based classification[J]. In Proceeding On IEEE Intelligent Vehicles Symposium,1998,143-148.
[40]H.A.P.Blom, E.A.Bloem. Joint immpda particle filter[C]. In Proceeding 6th International Conference on information Fusion volume 1, Cairns Queensland, 2003:785-792.
[41]J.Maccormick. Stochastic Algorithms for visual tracking[M]. New York:Springer-Verlag,2002.
[42]R.Morales-Men'endez, N.de Freitas, D.Poole. Real-time monitoring of complex industrial processes with particle filters[C]. In Proceeding of the Conference on Neural Information Processing Systems (NIPS), Vancover,2002:1433-1440.
[43]胡明昊,任明武,杨静宇．一种基于直方图模式的运动目标实时跟踪算法[J].计算机工程与应用,2004,3：71-74.
[44]O.Williams, A.Blake, R.Cipolla. A sparse probabilistic learning algorithm for real-time tracking[C]. In Proceedings of the Ninth IEEE International Conference on Computer Vision, Nice:IEEE Computer Society,2003:353-361.
[45]K.Toyama, A.Blake. Probabilistic tracking with exemplars in a metric space[J]. International Journal of Computer Vison,2002,48(1):9-19.