面向监视视频的运动轨迹提取方法研究
|
摘要
运动轨迹提取是智能视频监控的重要课题,涉及运动目标检测、目标分类、目标匹配等方面的内容。本文深入研究了监控场景中单个人体目标、多个人体目标的运动轨迹提取方法,解决了多目标部分遮挡情况下轨迹提取的难题。
轨迹提取要求确定每一帧视频中各个运动目标的位置,鉴于监控场景中运动目标的复杂性,本文提出了基于多元组的运动轨迹模型,不仅记录每帧视频中各个目标的位置信息,而且包含目标的运动方向、运动速度、外接矩形、轮廓参数以及是否发生遮挡等信息,丰富了轨迹概念的内涵,有利于进一步分析。
针对经典Camshift方法容易出现窗口漂移的问题,本文提出了一种基于时空约束策略的Camshift轨迹提取方法。该方法通过预测目标可能出现的位置约束搜索区域,根据运动检测结果对目标尺寸信息进行约束,使轨迹更接近目标实际情况。最后的仿真实验表明,该方法搜索速度快,轨迹提取精度高。
针对遮挡情况下多目标轨迹提取精度低的难题,本文提出了一种自适应区域分割的多目标运动轨迹提取方法,该方法能够根据目标是否存在遮挡情况自适应地切换目标匹配模式。仿真实验表明,该方法能够稳定地提取多个人体目标遮挡情况下的运动轨迹。
Motion trajectory extraction plays a very important role in video surveillance technology, which involves moving target detection, target classification, target matching, etc. In this thesis, the methods for extracting motion trajectories of single and multiple human targets for surveillance are researched and complicated problems in multi-target occluded scene are successfully solved.
In trajectory extraction, it is necessary to locate each target in each frame. Given the complicacy of moving targets in the surveillance scene, a trajectory extraction model based on multi-tuple is proposed in this thesis. In this model, not only is the position information recorded, but also the information of motion direction, speed, external rectangle, contour, and of whether the target is occluded is included. Therefore, the definition of trajectory is enriched in this model and it is conducive to further analysis.
A novel Camshift method based on spatiotemporal constraint strategy is proposed so as to solve the window-shift problem of the classic target tracking method-Camshift. In this method, the searching area is restricted by anticipating the possible position of target, and the results are more close to reality by restricting the information of size according to the results of motion detection. The results of simulate experiment demonstrate that both speed and precision are greatly improved.
Moreover, a trajectory extraction method of multiple targets is presented to solve the complicated problems of low precision in occluded multi-target motion trajectory extraction. The tracking modes can be adaptively shifted according to whether the target is obstructed or not. The results of simulate experiment show that the motion trajectory can be robustly extracted even though multiple human targets are occluded.
轨迹提取要求确定每一帧视频中各个运动目标的位置,鉴于监控场景中运动目标的复杂性,本文提出了基于多元组的运动轨迹模型,不仅记录每帧视频中各个目标的位置信息,而且包含目标的运动方向、运动速度、外接矩形、轮廓参数以及是否发生遮挡等信息,丰富了轨迹概念的内涵,有利于进一步分析。
针对经典Camshift方法容易出现窗口漂移的问题,本文提出了一种基于时空约束策略的Camshift轨迹提取方法。该方法通过预测目标可能出现的位置约束搜索区域,根据运动检测结果对目标尺寸信息进行约束,使轨迹更接近目标实际情况。最后的仿真实验表明,该方法搜索速度快,轨迹提取精度高。
针对遮挡情况下多目标轨迹提取精度低的难题,本文提出了一种自适应区域分割的多目标运动轨迹提取方法,该方法能够根据目标是否存在遮挡情况自适应地切换目标匹配模式。仿真实验表明,该方法能够稳定地提取多个人体目标遮挡情况下的运动轨迹。
Motion trajectory extraction plays a very important role in video surveillance technology, which involves moving target detection, target classification, target matching, etc. In this thesis, the methods for extracting motion trajectories of single and multiple human targets for surveillance are researched and complicated problems in multi-target occluded scene are successfully solved.
In trajectory extraction, it is necessary to locate each target in each frame. Given the complicacy of moving targets in the surveillance scene, a trajectory extraction model based on multi-tuple is proposed in this thesis. In this model, not only is the position information recorded, but also the information of motion direction, speed, external rectangle, contour, and of whether the target is occluded is included. Therefore, the definition of trajectory is enriched in this model and it is conducive to further analysis.
A novel Camshift method based on spatiotemporal constraint strategy is proposed so as to solve the window-shift problem of the classic target tracking method-Camshift. In this method, the searching area is restricted by anticipating the possible position of target, and the results are more close to reality by restricting the information of size according to the results of motion detection. The results of simulate experiment demonstrate that both speed and precision are greatly improved.
Moreover, a trajectory extraction method of multiple targets is presented to solve the complicated problems of low precision in occluded multi-target motion trajectory extraction. The tracking modes can be adaptively shifted according to whether the target is obstructed or not. The results of simulate experiment show that the motion trajectory can be robustly extracted even though multiple human targets are occluded.
引文
[1]孙即祥.图像处理[M].北京:科学出版社, 2009.
[2]李凤凯.多运动目标的检测与跟踪算法研究[J].天津大学. 2007.
[3] Dalal N, Triggs B. Histograms of oriented gradients for human detection[C]. Coference on Computer Vision and Pattern Recognition (CVPR). California, IEEE Computer Society Press. 2005, 1: 886-893.
[4] Dalal N. Finding people in Images and Videos[D]. Grenoble, France: The French National Institute for Research in Computer Science and Control, 2006.
[5]周柯.基于HOG特征的图像人体检测技术的研究与实现[D].武汉:华中科技大学, 2008.
[6]向国华.多视频目标跟踪算法研究[D].西安:西安电子科技大学, 2008.
[7]冯洋.多目标跟踪的数据关联算法研究[D].西安:西安电子科技大学, 2008.
[8]李凤凯.多运动目标的检测与跟踪算法研究[D].天津:天津大学, 2007.
[9]毛爽.多目标检测跟踪算法研究[D].济南:山东大学, 2008.
[10] WAN Y, WANG S. A Track-Before-Detect Algorithm Based on Extended H∞Particle Filter[J]. Chinese Journal of Electronics. 2011, 20(2): 375-379.
[11] YANG X, XING K, FENG X. Maneuvering Target Tracking in Dense Clutter Based on Particle Filtering[J]. Chinese Journal of Aeronautic. 2010, 24(2): 171-180.
[12] HAN G, MENG B. A new real time multi-target tracking algorithm in video [C]. International Conference on Computer Application and System Modeling(ICCASM 2010). 2010, 13: 568-572.
[13] Chen L, Guo B, Sun W. A New Multiple-Objects Tracking Method with Particle Filter[J]. 2009 Fifth International Conference on Information Assurance and Security. 2009: 281-284.
[14] Su H, SHUI P, LIU H, BAO Z. Particle Filter Based Track-before-detect Algorithm for Over-the-horizon Radar Target Detection and Tracking[J]. Chinese Journal of Electronics. 2009, 18(1): 59-64.
[15] Orton M, Fitzgerald W. A Bayesian approach to tracking multiple targets using sensor arrays and particle filters[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING. 2002, 50(2): 216-223.
[16] HUE C, CADRE J L, PEREZ P. Tracking multiple objects with particle filtering[J]. IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS. 2002, 38(3): 791-812.
[17]余弦.基于轨迹的视频运动对象的检测与跟踪技术研究[D].上海:上海交通大学, 2010.
[18] Ju S X, Black M J, Yacoob Y. Cardboard people: a parameterized model of articulated image motion[C]. Proceedings of the Second International Conference on Automatic Face and Gesture Recognition. 1996, 16: 38-44.
[19]王长军,朱善安.基于颜色和变形模板的实时人体检测[J].中国图象图形学报. 2006, 11(6): 861-866.
[20]王长军,朱善安.基于统计模型和活动轮廓的运动目标检测与跟踪[J].浙江大学学报(工学版). 2006, 40(2): 249-253.
[21]田丽丽.基于GVF Snake模型的轮廓提取方法研究[D].西安:陕西师范大学, 2009.
[22]金雪军.基于Snake模型的物体轮廓提取及跟踪技术研究[D].杭州:浙江工业大学, 2005.
[23]蔡珣,孟祥旭,刘强.一种新的基于区域的高速公路多车辆跟踪方案[J].光电工程. 2006, 27(3): 446-450.
[24]张乐华.多人跟踪中目标分裂与合并问题研究[J].北京邮电大学. 2009.
[25]陈怡君.利用轨迹特征分析行人异常行为[D].台湾:国立中央大学, 2006.
[26]刘闯.基于特征融合和MeanShift的多目标跟踪方法研究[J].苏州大学. 2009.
[27] Collins R, Lipton A, Kanade T. A System for Video Surveillance and Monitoring[R].Pittsburgh: Carnegie Mellon University, 2000.
[28] Haritaoglu I, Harwood D, Davis L. W4: Real-time surveillance of people and their activities[J]. IEEE Trans Pattern Analysis and Machine Intelligence. 2000, 22(8): 809-830.
[29]刘晓东,苏光大,周全.一种可视化智能户外监控系统[J].中国图象图形学报. 2000, 5(12): 1024-1029.
[30]万琴,王耀南.一种多运动目标检测跟踪方法研究与实现[J].计算机应用研究. 2007, 1: 199-202.
[31]郝久月,李超,高磊,熊璋.智能监控场景中运动目标轨迹聚类算法[J].北京航空航天大学学报. 2009, 35(9): 1083-1087.
[32]肖俊,庄越捉,吴飞.计算机视觉与机器学习技术在三维人体动画中的应用综述[J].计算机辅助设计与图形学学报. 2008, 20(3): 281-290.
[33]杨涛,李静,潘泉,张艳宁.一种基于贪心搜索的实时多目标遮挡处理算法[J].自动化学报. 2010, 36(3): 375-384.
[34]智利丁.视频监控中多运动对象的检测与跟踪[D].长沙:国防科技大学, 2007.
[35]李峥.智能视频监控中的遮挡目标跟踪技术研究[J].华中科技大学. 2008.
[36]李沛秦,谢剑斌,陈章永,程永茂,刘通.一种面向区域的快速去模糊算法[J].信号处理. 2010, 26(8): 1240-1245.
[37]谢剑斌,王晖,程江华,刘通.一种基于多重网格的光流场快速计算方法[J].仪器仪表学报. 2008, 29(4): 547-550.
[38]谢剑斌,闫玮,刘通,李沛秦.一种交通监视视频感兴趣区域提取算法[J].电视技术. 2008, 32(1): 82-84.
[39]秦陈刚.面向监视场景的斗殴行为检测技术研究[D].长沙:国防科技大学, 2010.
[40] Gonzalez R C, Woods R E. Digital Image Processing Second Edition:数字图像处理(第二版)[M].阮秋琦,阮宇智,译.北京:电子工业出版社, 2004.
[41]贾云得.机器视觉[M].北京:科学出版社, 2002.
[42]郭海峰.视频图像序列中运动目标检测与跟踪[D].长沙:国防科技大学, 2008.
[43]丁艳.基于全方位视觉的多目标检测跟踪[D].杭州:浙江大学, 2007.
[44]赵志旭.智能视觉监控中的运动目标检测与跟踪[D].重庆:电子科技大学, 2009.
[45] MIT Pedestrian Datasets. http://cbcl.mit.edu/software-datasets/PedestrianData. html.
[46] INRIA Database. http://lear.inrialpes.fr/data/human/.
[47] Porikli F. Integral histogram: A fast way to extract histograms in Cartesian spaces[C]. Proceedings of Coference on Computer Vision and Pattern Recognition. San Diego, California, USA, IEEE Computer Society Press. 2005, 1: 826-836.
[48]邓亚峰,苏光大,傅博.一种基于AdaBoost的快速动态人脸检测算法[J].计算机工程. 2006, 32(11): 222-224.
[49] Stojmenovic M. 6 Real time object detection in images based on an AdaBoost machine learning approach and a small training set[D]. Ottawa, Ontario: Ottawa-Carleton Institute for Computer Science School of Computer Science Carleton University, 2005.
[50]武勃,黄畅,艾海舟,劳世竑.基于连续Adaboost算法的多视角人脸检测[J].计算机研究与发展. 2005, 42(9): 1612-1621.
[51]王海川,张立明.一种新的Adaboost快速训练算法[J].复旦学报(自然科学版). 2004, 43(1): 27-33.
[52]赵江,徐鲁安.基于AdaBoost算法的目标检测[J].计算机工程. 2004, 30(4).
[53]吴立德.计算机视觉[M].上海:复旦大学出版社, 1993.
[54]王红梅,张科,李言俊.图像匹配研究进展[J].计算机工程与应用. 2004: 42-44, 77.
[55]卜凡艳.数字图像匹配技术研究[D].合肥:合肥工业大学, 2010.
[56]王军,张明柱.图像匹配算法的研究进展[J].大气与环境光学学报. 2007, 2(1): 11-15.
[57]熊凌.计算机视觉中的图像匹配综述[J].湖北工业大学学报. 2006, 21(3): 171-173.
[58]林志泉,胡永健,杨晖.一种基于特征的约束匹配方法[J].中国图象图形学报. 2007, 12(11): 2104-2108.
[59]陈皓,马彩文,陈岳承,孙小林,唐自力.基于灰度统计的快速模板匹配算法[J].光子学报. 2009, 38(6): 1586-1590.
[60]刘相滨,邹北骥,孙家广.基于边界跟踪的快速欧氏距离变换算法[J].计算机学报. 2006, 29(2): 317-323.
[61]吴成东,贾子熙,张云洲,黄月.基于欧氏距离的分布式网格定位估计方法[J].东北大学学报(自然科学版). 2009, 30(3): 325-328.
[62]诸葛婴,田捷,王蔚洪.三维欧氏距离变换的一种新方法[J].软件学报. 2001, 12(3): 383-389.
[63]郭志波,杨静宇,刘华军,严云洋.基于矩阵完备投影的快速主分量分析算法[J].中国图象图形学报. 2007, 12(4): 628-632.
[64]王安娜,张新华,谷召伟,潘博.基于改进Hausdorff测度的医学图像配准算法[J].电子学报. 2008, 36(11): 2247-2250.
[65]牛力丕,毛士艺,陈炜.基于Hausdorff距离的图像配准研究[J].电子与信息学报. 2007, 29(1): 35-38.
[66] Yang C H, Lai S H, Chang L W. Reliable image matching via modified hausdorff distance with normalized gradient consistency measure[C]. Proc of 3rd International Conference on Information Technology: Research and Education. Hsingchu, Taiwan, China, 2005, 158-161.
[67] Dubuisson M P, Jain A K. A modified hausdorff distance for object matching[C]. Proc of 12th IAPR International Conference on Pattern Recognition. Jerusalem, Israel, 1994, 566-568.
[68]申晓勇,雷英杰,蔡茹,张弛.基于加权Minkowski距离的IFS相异度度量方法[J].系统工程与电子技术. 2009, 31(6): 1358-1361.
[69]时永刚. Minkowski广义距离与多模态图像配准[J].北京理工大学学报. 2005, 25(10): 913-918.
[70]时永刚.新的Cauchy-Schwarz距离函数与多模态医学图像配准[J].光学技术. 2005, 31(5): 684-687, 690.
[71]董武.视频编码和编码转换中的运动矢量估计[D].合肥:安徽大学, 2007.
[72]谢琅,杨艳,曹阳,王立恒.一种改进的三步搜索块运动估计算法[J].武汉大学学报. 2005, 51(5): 625-628.
[73] Wang Y, Ostermann J, Zhang Y. Video Processing and Communications:视频处理与通信[M].王文全,侯正信,杨喜,译.北京:电子工业出版社, 2003.
[74] Barjatya A. Block Matching Algorithms for Motion Estimation[C]. DIP 6620 Spring 2004 Final Project Paper, Digital Image Processing. Utah State University, 2004.
[75]常发亮,马丽,乔谊正.视频序列中面向人的多目标跟踪算法[J].控制与决策. 2007, 22(4): 418-422.
[76]孙凯,刘士荣.多目标跟踪的改进Camshift-卡尔曼滤波组合算法[J].信息与控制. 2009, 28(1): 9-14.
[77] Wixson L, Hansen M. Detecting salient motion by accumulating directionally consistent flow[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2000, 22(8): 774-780.
[78] Bakowski A, Jones G A. Video surveillance tracking using colour region adjacency graphs[J]. IEE Conference Publication. 1999, 2(465): 794-798.
[79]李光,金学波.基于张量梯度直方图的MeanShift跟踪算法[J].浙江理工大学学报. 2009, 26(6): 880-885, 896.
[80]颜佳,吴敏渊,陈淑珍,张青林.跟踪窗口自适应的MeanShift跟踪[J].光学精密工程. 2009, 17(10): 2606-2611.
[81]初红霞,宋起超,王希凤,刘侠.基于帧间差分与Camshift相结合的目标跟踪算法[J].弹箭与制导学报. 2008, 28(3): 85-88.
[82] Fashing M, Tomasi C. Mean shift is a bound optimization[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2005, 27(3): 471-474.
[83] Fukunaga K, Hostetler L D. The estimation of gradient of a density function, with applications in pattern recognition[J]. IEEE Transactions On Information Theory. 1975, 21: 32-40.
[84] Bradski G R. Real time face and object tracking as a component of a perceptual user interface[C]. Proceeding of the 4th IEEE Workshop on Applications of Computer Vision. 1998, 214-219.
[85] Bradski G R. Computer video face tracking for use in a perceptual user interface[J]. Intel Technology Journal. 1998: 705-740.
[86] Cheng Y. Mean shift, mode seeking, and clustering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1995, 17(8): 790-799.
[87] Comaniciu D. Nonparametric information fusion for motion estimation[C]. IEEE International Conference on Computer Vision and Pattern Recognition. 2003, 1: 59-66.
[88] Comaniciu D. Mean Shift A Robust Approach Toward Feature Space Analysis[J]. IEEE Transactions on Pattern Analysis Machine Intelligence. 2002, 24(5): 603-619.
[89] Comaniciu D, Meer P. Mean shift analysis and application[C]. The proceeding of the 7th IEEE International Conference on Computer Vision. 1999, 2: 1197-1203.
[90] Bradski G, Kaebler A. Learning OpenCV:学习OpenCV(中文版)[M].于仕琪,刘瑞祯,译.北京:清华大学出版社, 2009.
[91]沈军行,孙守迁,潘云鹤.从运动捕获数据中提取关键帧[J].计算机辅助设计与图形学学报. 2004, 16(5): 719-723.
[92]罗忠祥,庄越挺,刘丰,潘云鹤.基于时空约束的运动编辑和运动重定向[J].计算机辅助设计与图形学学报. 2002, 14(12): 1146-1151.
[93]王江峰,张茂军,包卫东,熊志辉.双向时空连续性轨迹片段关联的目标跟踪方法[J].国防科技大学学报. 2011, 33(2): 44-48.
[94]王江峰,张茂军,熊志辉,包卫东.一种利用时空约束轨迹片段关联的目标跟踪方法[J].计算机应用研究. 2011, 28(3): 1165-1167, 1175.
[95] Sorenson H W. Kalman filtering: theory and application[M]. IEEE Press, 1985.
[96] PETS. http://www.filewatcher.com/b/ftp/ftp.cs.rdg.ac.uk/pub/.
[97] Koch C, Ullman S. Shifts in selective visual attention: Towards the underlying neuronal circuitry[J]. Human Neurobiology. 1985, 4: 219-227.
[98] Itti L, Koch C, Niebur E. A model of saliency-based visual attention for rapid scene analysis[J]. IEEE Trans on Pattern Analysis and Machine Intelligence. 1998, 20(11): 1254-1259.
[99]张焱,张志龙,沈振康,鹿小莺.基于动态显著性特征的粒子滤波多目标跟踪算法[J].电子学报. 2008, 36(12): 2305-2311.
[100]张国敏,殷建平,祝恩,毛玲.基于近似高斯金字塔的视觉注意模型快速算法[J].软件学报. 2009, 20(12): 3240-3253.
[101] Stentiford F W M. An evolutionary programming approach to the simulation of visual attention[C]. IEEE Proc of the IEEE Congress on Evolutionary Computation. Seoul, Korea, 2001, 851-858.
[102] Singer R A, Stein J J. An optimal tracking filter for processing sensor data of imprecisely determined origin in surveillance systems[C]. Proc of IEEE Conference on Decision and Control. Miami Beach, IEEE Press. 1971, 171-175.
[103] Mahler R. A theoretical foundation for the stein-winter probability hypothesis density (PHD) multi-target tracking approach[C]. Proc of the MSS National Symposium on Sensor and Data Fusion. 2000, 1(3): 99-118.
[2]李凤凯.多运动目标的检测与跟踪算法研究[J].天津大学. 2007.
[3] Dalal N, Triggs B. Histograms of oriented gradients for human detection[C]. Coference on Computer Vision and Pattern Recognition (CVPR). California, IEEE Computer Society Press. 2005, 1: 886-893.
[4] Dalal N. Finding people in Images and Videos[D]. Grenoble, France: The French National Institute for Research in Computer Science and Control, 2006.
[5]周柯.基于HOG特征的图像人体检测技术的研究与实现[D].武汉:华中科技大学, 2008.
[6]向国华.多视频目标跟踪算法研究[D].西安:西安电子科技大学, 2008.
[7]冯洋.多目标跟踪的数据关联算法研究[D].西安:西安电子科技大学, 2008.
[8]李凤凯.多运动目标的检测与跟踪算法研究[D].天津:天津大学, 2007.
[9]毛爽.多目标检测跟踪算法研究[D].济南:山东大学, 2008.
[10] WAN Y, WANG S. A Track-Before-Detect Algorithm Based on Extended H∞Particle Filter[J]. Chinese Journal of Electronics. 2011, 20(2): 375-379.
[11] YANG X, XING K, FENG X. Maneuvering Target Tracking in Dense Clutter Based on Particle Filtering[J]. Chinese Journal of Aeronautic. 2010, 24(2): 171-180.
[12] HAN G, MENG B. A new real time multi-target tracking algorithm in video [C]. International Conference on Computer Application and System Modeling(ICCASM 2010). 2010, 13: 568-572.
[13] Chen L, Guo B, Sun W. A New Multiple-Objects Tracking Method with Particle Filter[J]. 2009 Fifth International Conference on Information Assurance and Security. 2009: 281-284.
[14] Su H, SHUI P, LIU H, BAO Z. Particle Filter Based Track-before-detect Algorithm for Over-the-horizon Radar Target Detection and Tracking[J]. Chinese Journal of Electronics. 2009, 18(1): 59-64.
[15] Orton M, Fitzgerald W. A Bayesian approach to tracking multiple targets using sensor arrays and particle filters[J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING. 2002, 50(2): 216-223.
[16] HUE C, CADRE J L, PEREZ P. Tracking multiple objects with particle filtering[J]. IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS. 2002, 38(3): 791-812.
[17]余弦.基于轨迹的视频运动对象的检测与跟踪技术研究[D].上海:上海交通大学, 2010.
[18] Ju S X, Black M J, Yacoob Y. Cardboard people: a parameterized model of articulated image motion[C]. Proceedings of the Second International Conference on Automatic Face and Gesture Recognition. 1996, 16: 38-44.
[19]王长军,朱善安.基于颜色和变形模板的实时人体检测[J].中国图象图形学报. 2006, 11(6): 861-866.
[20]王长军,朱善安.基于统计模型和活动轮廓的运动目标检测与跟踪[J].浙江大学学报(工学版). 2006, 40(2): 249-253.
[21]田丽丽.基于GVF Snake模型的轮廓提取方法研究[D].西安:陕西师范大学, 2009.
[22]金雪军.基于Snake模型的物体轮廓提取及跟踪技术研究[D].杭州:浙江工业大学, 2005.
[23]蔡珣,孟祥旭,刘强.一种新的基于区域的高速公路多车辆跟踪方案[J].光电工程. 2006, 27(3): 446-450.
[24]张乐华.多人跟踪中目标分裂与合并问题研究[J].北京邮电大学. 2009.
[25]陈怡君.利用轨迹特征分析行人异常行为[D].台湾:国立中央大学, 2006.
[26]刘闯.基于特征融合和MeanShift的多目标跟踪方法研究[J].苏州大学. 2009.
[27] Collins R, Lipton A, Kanade T. A System for Video Surveillance and Monitoring[R].Pittsburgh: Carnegie Mellon University, 2000.
[28] Haritaoglu I, Harwood D, Davis L. W4: Real-time surveillance of people and their activities[J]. IEEE Trans Pattern Analysis and Machine Intelligence. 2000, 22(8): 809-830.
[29]刘晓东,苏光大,周全.一种可视化智能户外监控系统[J].中国图象图形学报. 2000, 5(12): 1024-1029.
[30]万琴,王耀南.一种多运动目标检测跟踪方法研究与实现[J].计算机应用研究. 2007, 1: 199-202.
[31]郝久月,李超,高磊,熊璋.智能监控场景中运动目标轨迹聚类算法[J].北京航空航天大学学报. 2009, 35(9): 1083-1087.
[32]肖俊,庄越捉,吴飞.计算机视觉与机器学习技术在三维人体动画中的应用综述[J].计算机辅助设计与图形学学报. 2008, 20(3): 281-290.
[33]杨涛,李静,潘泉,张艳宁.一种基于贪心搜索的实时多目标遮挡处理算法[J].自动化学报. 2010, 36(3): 375-384.
[34]智利丁.视频监控中多运动对象的检测与跟踪[D].长沙:国防科技大学, 2007.
[35]李峥.智能视频监控中的遮挡目标跟踪技术研究[J].华中科技大学. 2008.
[36]李沛秦,谢剑斌,陈章永,程永茂,刘通.一种面向区域的快速去模糊算法[J].信号处理. 2010, 26(8): 1240-1245.
[37]谢剑斌,王晖,程江华,刘通.一种基于多重网格的光流场快速计算方法[J].仪器仪表学报. 2008, 29(4): 547-550.
[38]谢剑斌,闫玮,刘通,李沛秦.一种交通监视视频感兴趣区域提取算法[J].电视技术. 2008, 32(1): 82-84.
[39]秦陈刚.面向监视场景的斗殴行为检测技术研究[D].长沙:国防科技大学, 2010.
[40] Gonzalez R C, Woods R E. Digital Image Processing Second Edition:数字图像处理(第二版)[M].阮秋琦,阮宇智,译.北京:电子工业出版社, 2004.
[41]贾云得.机器视觉[M].北京:科学出版社, 2002.
[42]郭海峰.视频图像序列中运动目标检测与跟踪[D].长沙:国防科技大学, 2008.
[43]丁艳.基于全方位视觉的多目标检测跟踪[D].杭州:浙江大学, 2007.
[44]赵志旭.智能视觉监控中的运动目标检测与跟踪[D].重庆:电子科技大学, 2009.
[45] MIT Pedestrian Datasets. http://cbcl.mit.edu/software-datasets/PedestrianData. html.
[46] INRIA Database. http://lear.inrialpes.fr/data/human/.
[47] Porikli F. Integral histogram: A fast way to extract histograms in Cartesian spaces[C]. Proceedings of Coference on Computer Vision and Pattern Recognition. San Diego, California, USA, IEEE Computer Society Press. 2005, 1: 826-836.
[48]邓亚峰,苏光大,傅博.一种基于AdaBoost的快速动态人脸检测算法[J].计算机工程. 2006, 32(11): 222-224.
[49] Stojmenovic M. 6 Real time object detection in images based on an AdaBoost machine learning approach and a small training set[D]. Ottawa, Ontario: Ottawa-Carleton Institute for Computer Science School of Computer Science Carleton University, 2005.
[50]武勃,黄畅,艾海舟,劳世竑.基于连续Adaboost算法的多视角人脸检测[J].计算机研究与发展. 2005, 42(9): 1612-1621.
[51]王海川,张立明.一种新的Adaboost快速训练算法[J].复旦学报(自然科学版). 2004, 43(1): 27-33.
[52]赵江,徐鲁安.基于AdaBoost算法的目标检测[J].计算机工程. 2004, 30(4).
[53]吴立德.计算机视觉[M].上海:复旦大学出版社, 1993.
[54]王红梅,张科,李言俊.图像匹配研究进展[J].计算机工程与应用. 2004: 42-44, 77.
[55]卜凡艳.数字图像匹配技术研究[D].合肥:合肥工业大学, 2010.
[56]王军,张明柱.图像匹配算法的研究进展[J].大气与环境光学学报. 2007, 2(1): 11-15.
[57]熊凌.计算机视觉中的图像匹配综述[J].湖北工业大学学报. 2006, 21(3): 171-173.
[58]林志泉,胡永健,杨晖.一种基于特征的约束匹配方法[J].中国图象图形学报. 2007, 12(11): 2104-2108.
[59]陈皓,马彩文,陈岳承,孙小林,唐自力.基于灰度统计的快速模板匹配算法[J].光子学报. 2009, 38(6): 1586-1590.
[60]刘相滨,邹北骥,孙家广.基于边界跟踪的快速欧氏距离变换算法[J].计算机学报. 2006, 29(2): 317-323.
[61]吴成东,贾子熙,张云洲,黄月.基于欧氏距离的分布式网格定位估计方法[J].东北大学学报(自然科学版). 2009, 30(3): 325-328.
[62]诸葛婴,田捷,王蔚洪.三维欧氏距离变换的一种新方法[J].软件学报. 2001, 12(3): 383-389.
[63]郭志波,杨静宇,刘华军,严云洋.基于矩阵完备投影的快速主分量分析算法[J].中国图象图形学报. 2007, 12(4): 628-632.
[64]王安娜,张新华,谷召伟,潘博.基于改进Hausdorff测度的医学图像配准算法[J].电子学报. 2008, 36(11): 2247-2250.
[65]牛力丕,毛士艺,陈炜.基于Hausdorff距离的图像配准研究[J].电子与信息学报. 2007, 29(1): 35-38.
[66] Yang C H, Lai S H, Chang L W. Reliable image matching via modified hausdorff distance with normalized gradient consistency measure[C]. Proc of 3rd International Conference on Information Technology: Research and Education. Hsingchu, Taiwan, China, 2005, 158-161.
[67] Dubuisson M P, Jain A K. A modified hausdorff distance for object matching[C]. Proc of 12th IAPR International Conference on Pattern Recognition. Jerusalem, Israel, 1994, 566-568.
[68]申晓勇,雷英杰,蔡茹,张弛.基于加权Minkowski距离的IFS相异度度量方法[J].系统工程与电子技术. 2009, 31(6): 1358-1361.
[69]时永刚. Minkowski广义距离与多模态图像配准[J].北京理工大学学报. 2005, 25(10): 913-918.
[70]时永刚.新的Cauchy-Schwarz距离函数与多模态医学图像配准[J].光学技术. 2005, 31(5): 684-687, 690.
[71]董武.视频编码和编码转换中的运动矢量估计[D].合肥:安徽大学, 2007.
[72]谢琅,杨艳,曹阳,王立恒.一种改进的三步搜索块运动估计算法[J].武汉大学学报. 2005, 51(5): 625-628.
[73] Wang Y, Ostermann J, Zhang Y. Video Processing and Communications:视频处理与通信[M].王文全,侯正信,杨喜,译.北京:电子工业出版社, 2003.
[74] Barjatya A. Block Matching Algorithms for Motion Estimation[C]. DIP 6620 Spring 2004 Final Project Paper, Digital Image Processing. Utah State University, 2004.
[75]常发亮,马丽,乔谊正.视频序列中面向人的多目标跟踪算法[J].控制与决策. 2007, 22(4): 418-422.
[76]孙凯,刘士荣.多目标跟踪的改进Camshift-卡尔曼滤波组合算法[J].信息与控制. 2009, 28(1): 9-14.
[77] Wixson L, Hansen M. Detecting salient motion by accumulating directionally consistent flow[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2000, 22(8): 774-780.
[78] Bakowski A, Jones G A. Video surveillance tracking using colour region adjacency graphs[J]. IEE Conference Publication. 1999, 2(465): 794-798.
[79]李光,金学波.基于张量梯度直方图的MeanShift跟踪算法[J].浙江理工大学学报. 2009, 26(6): 880-885, 896.
[80]颜佳,吴敏渊,陈淑珍,张青林.跟踪窗口自适应的MeanShift跟踪[J].光学精密工程. 2009, 17(10): 2606-2611.
[81]初红霞,宋起超,王希凤,刘侠.基于帧间差分与Camshift相结合的目标跟踪算法[J].弹箭与制导学报. 2008, 28(3): 85-88.
[82] Fashing M, Tomasi C. Mean shift is a bound optimization[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2005, 27(3): 471-474.
[83] Fukunaga K, Hostetler L D. The estimation of gradient of a density function, with applications in pattern recognition[J]. IEEE Transactions On Information Theory. 1975, 21: 32-40.
[84] Bradski G R. Real time face and object tracking as a component of a perceptual user interface[C]. Proceeding of the 4th IEEE Workshop on Applications of Computer Vision. 1998, 214-219.
[85] Bradski G R. Computer video face tracking for use in a perceptual user interface[J]. Intel Technology Journal. 1998: 705-740.
[86] Cheng Y. Mean shift, mode seeking, and clustering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1995, 17(8): 790-799.
[87] Comaniciu D. Nonparametric information fusion for motion estimation[C]. IEEE International Conference on Computer Vision and Pattern Recognition. 2003, 1: 59-66.
[88] Comaniciu D. Mean Shift A Robust Approach Toward Feature Space Analysis[J]. IEEE Transactions on Pattern Analysis Machine Intelligence. 2002, 24(5): 603-619.
[89] Comaniciu D, Meer P. Mean shift analysis and application[C]. The proceeding of the 7th IEEE International Conference on Computer Vision. 1999, 2: 1197-1203.
[90] Bradski G, Kaebler A. Learning OpenCV:学习OpenCV(中文版)[M].于仕琪,刘瑞祯,译.北京:清华大学出版社, 2009.
[91]沈军行,孙守迁,潘云鹤.从运动捕获数据中提取关键帧[J].计算机辅助设计与图形学学报. 2004, 16(5): 719-723.
[92]罗忠祥,庄越挺,刘丰,潘云鹤.基于时空约束的运动编辑和运动重定向[J].计算机辅助设计与图形学学报. 2002, 14(12): 1146-1151.
[93]王江峰,张茂军,包卫东,熊志辉.双向时空连续性轨迹片段关联的目标跟踪方法[J].国防科技大学学报. 2011, 33(2): 44-48.
[94]王江峰,张茂军,熊志辉,包卫东.一种利用时空约束轨迹片段关联的目标跟踪方法[J].计算机应用研究. 2011, 28(3): 1165-1167, 1175.
[95] Sorenson H W. Kalman filtering: theory and application[M]. IEEE Press, 1985.
[96] PETS. http://www.filewatcher.com/b/ftp/ftp.cs.rdg.ac.uk/pub/.
[97] Koch C, Ullman S. Shifts in selective visual attention: Towards the underlying neuronal circuitry[J]. Human Neurobiology. 1985, 4: 219-227.
[98] Itti L, Koch C, Niebur E. A model of saliency-based visual attention for rapid scene analysis[J]. IEEE Trans on Pattern Analysis and Machine Intelligence. 1998, 20(11): 1254-1259.
[99]张焱,张志龙,沈振康,鹿小莺.基于动态显著性特征的粒子滤波多目标跟踪算法[J].电子学报. 2008, 36(12): 2305-2311.
[100]张国敏,殷建平,祝恩,毛玲.基于近似高斯金字塔的视觉注意模型快速算法[J].软件学报. 2009, 20(12): 3240-3253.
[101] Stentiford F W M. An evolutionary programming approach to the simulation of visual attention[C]. IEEE Proc of the IEEE Congress on Evolutionary Computation. Seoul, Korea, 2001, 851-858.
[102] Singer R A, Stein J J. An optimal tracking filter for processing sensor data of imprecisely determined origin in surveillance systems[C]. Proc of IEEE Conference on Decision and Control. Miami Beach, IEEE Press. 1971, 171-175.
[103] Mahler R. A theoretical foundation for the stein-winter probability hypothesis density (PHD) multi-target tracking approach[C]. Proc of the MSS National Symposium on Sensor and Data Fusion. 2000, 1(3): 99-118.