动态场景下运动目标检测与跟踪
|
摘要
运动目标检测和跟踪一直是计算机视觉领域内研究的热点问题,它在智能监控,武器装备,自主导航车,医学诊断等众多领域都有广泛的应用。而动态场景下运动目标检测和跟踪由于摄像机的运动而造成背景的不断变化,很多问题如目标的准确检测,目标形变跟踪及遮挡问题至今尚未解决。于是,本文将针对动态场景下中所遇到的一些问题进行研究,主要工作可以归纳如下：
1针对摄像头的运动而造成的背景不断变化且对不同层次上的物体所造成的运动向量也不同,提出动态场景下基于对极几何约束判别静与动的点的基础上检测运动目标,即前一帧中静态的特征点一定存在于相继帧中对应的极线上,如果远离对应的极线,则判定为动态的点。从而在背景景物差异大情况下依然能够检测出运动目标。
2针对传统的SIFT算法中特征描述子计算复杂度高的情况,提出以积分直方图快速构建特征描述子的方法,降低其计算的复杂性,使其能够满足目标检测和跟踪时的实时性。
3针对动态场景下中的运动目标经常会发生尺度变化,在分析传统均值漂移算法核带宽固定的基础上,采用改进的自适应尺度变化的均值漂移算法。
4动态场景下运动目标会经常遇到静态障碍物出现遮挡或者多个动态目标运动过程中粘贴存在相互遮挡等情况,于是,针对动态目标遇到静态障碍物遮挡时,采用结合卡尔曼滤波估计的均值漂移算法来估计在遇到遮挡情况下目标的位置。并针对在摄像机运动下,两个或多个目标的存在相互遮挡且这些目标形状尺度同时也在变化时,提出结合实时检测的运动目标跟踪算法,将实时的检测结果更新到跟踪算法中,并将跟踪的结果反馈检测的准确性。
最后,对本文的研究工作进行了总结,并提出了后续研究工作的思路,对今后的研究具有一定的指导意义。
Moving object detection and tracking remain as the hot spot issue in the field of computer vision. They have been widely used in various fields like intelligent monitoring, military hardware, autonomous navigation and medical diagnosis. However, Many questions have not been resolved yet in the dynamic scenes of moving object detection and tracking because of the flux and reflux of the moving camera. Thus the paper conducts an investigation to the encountered problem in dynamic scenes. The main tasks can be summed up as follows:
1. To sovle the problem of the constantly changing background caused by the moving camera and the different motion vector caused by objects of different level, We propose geometric constraints method based on SIFT to classify a pixel as moving or static. The constraint, we use, is the multi-view epipolar constraint which requires images of static points to lie on the corresponding epipolar lines in subsequent images. If the point far away from the corresponding epipolar line, then we can set the point as a static point. Thereby, moving objects can be detected even in the case of widely difference of the background scenes.
2. Aimed at the case of high computation complexity of feature descriptor in the conventional SIFT algorithm, we propose the method of rapid building feature descriptor using the integral histogram, which is to reduce the computation complexity and satisfy the real-time of object detection and tracing.
3. Aimed at the scale changing of the moving objects in the dynamic scene, we propose the improved mean shift algorithm of self-adapting changing scale on the basis of analyzing the fixed kernel bandwidth of conventional mean shift algorithm.
4. The moving objects frequently meet the case of static barrier shading or the several dynamic targets pasting and shading with each other in the motor process. Accordingly, aimed at the static barrier shading of dynamic objects, we propose the mean shift algorithm combined with the Kallman filtering to estimate the target location in the case of shading. And aimed at two or more dynamic objects shading with each other and their shape and scale changing in the case of camera moving, we propose the moving objects tracing algorithm combined with the real time detection and update the tracing algorithm of the real time detection result. The tracing result feeds back the detection veracity.
At last, the thesis sums up the research work and raises the train of thought of the follow-up research work. It will be have directive significance to the future research.
1针对摄像头的运动而造成的背景不断变化且对不同层次上的物体所造成的运动向量也不同,提出动态场景下基于对极几何约束判别静与动的点的基础上检测运动目标,即前一帧中静态的特征点一定存在于相继帧中对应的极线上,如果远离对应的极线,则判定为动态的点。从而在背景景物差异大情况下依然能够检测出运动目标。
2针对传统的SIFT算法中特征描述子计算复杂度高的情况,提出以积分直方图快速构建特征描述子的方法,降低其计算的复杂性,使其能够满足目标检测和跟踪时的实时性。
3针对动态场景下中的运动目标经常会发生尺度变化,在分析传统均值漂移算法核带宽固定的基础上,采用改进的自适应尺度变化的均值漂移算法。
4动态场景下运动目标会经常遇到静态障碍物出现遮挡或者多个动态目标运动过程中粘贴存在相互遮挡等情况,于是,针对动态目标遇到静态障碍物遮挡时,采用结合卡尔曼滤波估计的均值漂移算法来估计在遇到遮挡情况下目标的位置。并针对在摄像机运动下,两个或多个目标的存在相互遮挡且这些目标形状尺度同时也在变化时,提出结合实时检测的运动目标跟踪算法,将实时的检测结果更新到跟踪算法中,并将跟踪的结果反馈检测的准确性。
最后,对本文的研究工作进行了总结,并提出了后续研究工作的思路,对今后的研究具有一定的指导意义。
Moving object detection and tracking remain as the hot spot issue in the field of computer vision. They have been widely used in various fields like intelligent monitoring, military hardware, autonomous navigation and medical diagnosis. However, Many questions have not been resolved yet in the dynamic scenes of moving object detection and tracking because of the flux and reflux of the moving camera. Thus the paper conducts an investigation to the encountered problem in dynamic scenes. The main tasks can be summed up as follows:
1. To sovle the problem of the constantly changing background caused by the moving camera and the different motion vector caused by objects of different level, We propose geometric constraints method based on SIFT to classify a pixel as moving or static. The constraint, we use, is the multi-view epipolar constraint which requires images of static points to lie on the corresponding epipolar lines in subsequent images. If the point far away from the corresponding epipolar line, then we can set the point as a static point. Thereby, moving objects can be detected even in the case of widely difference of the background scenes.
2. Aimed at the case of high computation complexity of feature descriptor in the conventional SIFT algorithm, we propose the method of rapid building feature descriptor using the integral histogram, which is to reduce the computation complexity and satisfy the real-time of object detection and tracing.
3. Aimed at the scale changing of the moving objects in the dynamic scene, we propose the improved mean shift algorithm of self-adapting changing scale on the basis of analyzing the fixed kernel bandwidth of conventional mean shift algorithm.
4. The moving objects frequently meet the case of static barrier shading or the several dynamic targets pasting and shading with each other in the motor process. Accordingly, aimed at the static barrier shading of dynamic objects, we propose the mean shift algorithm combined with the Kallman filtering to estimate the target location in the case of shading. And aimed at two or more dynamic objects shading with each other and their shape and scale changing in the case of camera moving, we propose the moving objects tracing algorithm combined with the real time detection and update the tracing algorithm of the real time detection result. The tracing result feeds back the detection veracity.
At last, the thesis sums up the research work and raises the train of thought of the follow-up research work. It will be have directive significance to the future research.
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]Marr D, Vision, Freeman W.H.and Company, San Francisico,1982.
[3]P KaewTrakulPong, R Bowden. A real time adaptive visual surveillance system for tracking low-resolution colour targets in dynamically changing scenes[J]. Image and Vision Computing,2003,21(10):913-929.
[4]T Zhao, R Nevatia. Tracking multiple humans in complex situations[J]. IEEE Tra-nsactions on Pattern Analysis and Machine Intelligence,2004,26(9):1208-1221.
[5]Q Z Li, D X He, B Wang. Effective Moving Objects Detection Based on Cluster-ing Background Model for Video Surveillance[J]. Image and signal Processing, CISP'08. Congress 011,2008(3):656-660.
[6]I Pavlidis, V Morellas, P Tsiamyrtzis, Harp S. Urban surveillance system:From the laboratory to the commercial world[J]. Proceedings of the IEEE,2001,89(10): 1478-1497.
[7]B Jung, G S Sukhatme. A generalized region-based approach for multi-target tracking in outdoor environments[C]. In:Proceedings of IEEE International Conference on Robotics and Automation, New Orleans, LA, United States,2004: 2189-2195.
[8]B Jung, G S Sukhatme. Detecting moving objects using a single camera on a mobile robot in an outdoor environment[C]. In:Proceedings of International Conference on Intelligent Autonomous Systems,2004:980-987.
[9]H Nait-Charif, S J McKenna. Head tracking and action recognition in a smart meeting room[C]. In:Proceedings of 4th IEEE International Workshop on Perfor mance Evaluation of Tracking and Surveillance, Graz, Austria,2003:24-31.
[10]Y Wu, T S Huang. Robust visual tracking by integrating multiple cues based on co-inference learning[J]. International Journal of Computer Vision,2004,58(1): 55-71.
[11]曹银花,李林,郜广军,等.动摄像机和动目标跟踪模式下的目标检测新方法[J].光学技术,2005,31(2)：276-278.
[12]邵文坤,黄爱民,韦庆.动态场景下的运动目标跟踪方法研究[J].计算机仿真,2006,23(5)：181-184.
[13]郑世民.动态场景图像序列中运动目标检测与跟踪：[博士学位论文].南京：东南大学,2005.1-12.
[14]S Lampotang, D E, Lizdas J, et al. Panoramic screen-based simulation with dynamic background. Simulation Conference[C], WSC2008. Winter:2947-2947.
[15]R Collins, A Lipton. A system for video surveillance and monitoring:VSAM final report. CMU-RI-TR-00-12[J]. Robotic Institute Carnegie Mellon University. 2000.
[16]王亮,胡卫明,谭铁牛.人运动的视觉分析综述[J].计算机学报.2002(3)：1-16.
[17]V Mahadevan, N Vasconcelos. Background subtraction in highly dynamic sce-nes. Computer Vision and Pattern Recognition[C], IEEE Conference on CVPR. 2008:1-6.
[18]I Visentini. On Stable Dynamic Background Generation Technique Using Gaussian Mixture Models for Robust Object Detection. Advanced Video and Signal Based Surveillance[C], IEEE International Conference on AVSS'08.2008: 41-48.
[19]L Liu, Y J Wang. Robotic Dynamic Target Recognition and Tracking Based on the Monoeular Vision[C]. Control Conference, Chinese.2008:193-197
[20]Lee J S, Rhee K Y, Kim S D. Moving Target Tracking Algorithm Based on the Confidence Measure of Motion Vectors[C]. Proc of the IEEE In the Conf on Image Processing,2001:369-372.
[21]Thakoor N, Gao J. Automatic Video Object Shape Extraction and Its Classifica-tion with Camera in Motion[C]. Proccedings of IEEE In the Conf on Image Processing,2005:437-440.
[22]郑世友,费树岷.动态场景图像序列中运动目标检测新方法[C].中国图象图形学报2007(12)：1590-1597.
[23]Z Chen, S T Birchfield. Person Following with a MobileRobot Using Binocular Feature-Based Tracking[J]. IROS,2007:815-820.
[24]Wang L, Hu W M, Tan T N. A survey of visual analysis of human motion[J]. Chinese Journal of Computers,2002,25(3):225-237.
[25]Jepson A, Fleet D, EI-Maraghi T. Robust online appearance models for visual tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003,25(10):1296-1311.
[26]Matthews I, Ishikawa T, Baker S. The template update problem[J]. IEEE Transa-ctionson Pattern Analysis and Machine Intelligence,2004,26(6):810-815.
[27]Luo H, Eleftheriadis. A Model-based segmentation and tracking of head-and-shouder video objects for real time multimedia services[J]. IEEE Transactions on Multimedia,2003,5(3):379-389.
[28]Kato J, Watanabe T, Joga S, et al. An HMM/MRF-based stochastic framework for robust vehicle tracking[J]. IEEE Transactions on Intelligent Transportation Sys-terns.2004,5(3):142-154.
[29]Djuric P M, Kotecha J H, ZhangJ, et al. Particle filtering[J]. IEEE Signal Process-ing Magazine,2003,20(5):19-38.
[30]Gastand M, Barlaud M, Aubert G Combining shape Prior and statistical features for active contour segnlentation[J]. IEEE Transactions on Circuits and Systems for Video Technology,2004,14(5):726-734.
[31]Kap-hos, Ju-jang L. Object tracking using adaptive color Snake model[C]. in: Proeeedings 2003 IEEE/ASME International Conference on Advanced Intelligent Mechattronies(AIM2003),2003.1406-1410.
[32]Kim W, Lee J. Visual tracking using snake based on targets contour information [C]. in:IEEE International SymPosium on Industrial Eleetronies. Pusan. Institute of Eleetrieal and Electronies Engineers Inc,2001:176-180.
[33]V Mahadevan, N Vasconcelos. Background subtraction in highly dynamic seen-es[C]. Computer Vision and Pattern Recognition, CVPR 2008. IEEE Conference on,2008:1-6.
[34]S Lampotang, D E Lizdas, J J Tumino. Panoramic screen-based simulation with dynamic background[C]. Simulation Conference, Winter,2008:2937-2947.
[35]A J Lipton, H Fujiyoshi, R S Patil. Moving Target Classification and Tracking from Real-time Video[C], in Proc. of the 4th IEEE Workshop on Application of Computer Vision,1998:8-14.
[36]M Piaggio, P Fornaro, A Piombo, et al. An optical-flow person following behave-iour[C], In IEEE ISIC/CIRNISAS Joint Conference,1998:4078-4083.
[37]G Chivil'o, F Mezzaro, A Sgorbissa, et al, Follow the leader behaviour through optical flow minimization[C], IEEE International Conference on Intelligent Robots and Systems IROS,2004(4):3182-3187.
[38]A Handa, J Sivaswamy, K M Krishna, et al. Person Following with a Mobile Robot Using a Modified Optical Flow[J], Advances in Mobile Robotics, World Scientific,2008:1154-1160
[39]Heuer J, Kaup A. Global motion estimation in image sequences using robust motion vector field segmentation[C], In:Proceeding of the 7th ACM internation-al conference on Multimedia, Sydney, Australia,1999:261-264.
[40]Cheung H K, Siu W C. Fast global motion estimation for sprite generation[J], In: Proceeding of the IEEE International Symposium on Circuits and Systems (ISCAS) Arizona, USA,2002(3):26-29.
[41]Sorwar G, Mursher M, Dooley L. Fast global motion estimation using iterative least-square technique[C], In:Proceedings of the 4th International Conference on Information Communications&Signal, Proceedings of the 4th IEEE Pacific-Rim Conference on Multimedia, Singapore,2003(1):282-286.
[42]D G Lowe. Distinctive Image Features from Scale Invariant Keypoints[J], Intern-ational Journal of Computer Vision,2004,60(2):91-111.
[43]D G Lowe. Local Feature View Clustering for 3D Object Recognition[J]. IEEE Conference on Computer Vision and Pattern Recognition, Kauai, Hawaii,2001(1): 682-688.
[44]T Lindberg. Scale-Space for discrete Signals[J]. IEEE Trans. PAMI,1990(12): 234-254.
[45]M Brown, D G Lowe, Invariant Features from Interest Point Groups[C]. British Machine Vision Conference, BMVC 2002, Cardiff,Wales:British Machine Vision Association,2002:656-665.
[46]F Crow. Summed-area tables for texture mapping[J]. Proceedings of SIGGRAPH, 1984,18(3):207-212.
[47]P Viola, M Jones. Rapid object detection using a boosted cascade of simple features[J]. CVPR,2001(1):511-518.
[48]Hartley R Zisserman. A Multiple view geometry in computer vision[M]. Cambri-dge University Press 2000.
[49]K Inui, S Kaneko, S Igarashi. Robust line fitting using LmedS clustering, Syste-ms and Computers in Japan[J].2003,34:92-100.
[50]Fischler M, Bolles R. Random sample consensus:A paradigm for model fitting with applications to image analysis and automated cartography[J], Communi-caions of the ACM,1981(24):381-385. -caions of the ACM,1981(24):381-385.
[51]S P Lloyd. Least Squares Quantization in PCM, IEEE Transactions on Inform-ation Theory 28[J].1982:129-137.
[52]Keinosuke Fukunage, Larry D Hosterler. The estimation of the gradient of a density function, with applications in pattern recognition[J]. IEEE Transactions oninformation theory, January 1975(1):5-20.
[53]Yizong Cheng. Mean Shift, mode seeking and clustering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995:8-17.
[54]Comaniciu D. Nonparametric information fusion for motion estimation[C]. In: Proc.Of the IEEE Conf.on Computer Vision and Pattern Recognition (CVPR). 2003:59-66.
[55]Comaniciu D, Ramesh V, Bue AD.Multivariate saddle point detection for statistic-cal clustering[C]. In:Proc.of the European Confence on Computer Vision (ECCV).2002:23-26.
[56]A Bhattacharyya. On a measure of divergence between two statistical populations defined by their probability distributions[J]. Bull.Calcutta Math,1943:99-109.
[57]左军毅,梁彦,赵春晖.Mean Shift跟踪算法中尺度自适应策略的研究[J].中国图象图形学报,2008,13(9)：1750-1757.
[58]Yilmaz A. Object Tracking by Asymmetric Kernel Mean Shift will Automatic Scale and Orientation Selection[C]. IEEE Conf. On Computer Vision and Pattern Recognition,2007:1-6.
[59]COMANICIU D, RAMESH V, MEER P. Kernel-based object tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(5):564-577.
[60]R E Kalman, A new approach to linear filtering and prediction problems[J], Journal of Basic Engineering 1960:35-45.
[61]Gregory D Halter, Peter N Belbumeur.Efficient Region Tracking With Parametric Models of Geometry and Illumination[J]. IEEE Transactions On Pattern Analysis And Machine Intelligence, October 1998(10):1025-1038.
[62]S Th run, W Burgard, and D Fox, Probabilistic Robotics:Intelligent Robotics and Autonomus Agents[M], Cambridge, MA:MIT Press,2005.
[2]Marr D, Vision, Freeman W.H.and Company, San Francisico,1982.
[3]P KaewTrakulPong, R Bowden. A real time adaptive visual surveillance system for tracking low-resolution colour targets in dynamically changing scenes[J]. Image and Vision Computing,2003,21(10):913-929.
[4]T Zhao, R Nevatia. Tracking multiple humans in complex situations[J]. IEEE Tra-nsactions on Pattern Analysis and Machine Intelligence,2004,26(9):1208-1221.
[5]Q Z Li, D X He, B Wang. Effective Moving Objects Detection Based on Cluster-ing Background Model for Video Surveillance[J]. Image and signal Processing, CISP'08. Congress 011,2008(3):656-660.
[6]I Pavlidis, V Morellas, P Tsiamyrtzis, Harp S. Urban surveillance system:From the laboratory to the commercial world[J]. Proceedings of the IEEE,2001,89(10): 1478-1497.
[7]B Jung, G S Sukhatme. A generalized region-based approach for multi-target tracking in outdoor environments[C]. In:Proceedings of IEEE International Conference on Robotics and Automation, New Orleans, LA, United States,2004: 2189-2195.
[8]B Jung, G S Sukhatme. Detecting moving objects using a single camera on a mobile robot in an outdoor environment[C]. In:Proceedings of International Conference on Intelligent Autonomous Systems,2004:980-987.
[9]H Nait-Charif, S J McKenna. Head tracking and action recognition in a smart meeting room[C]. In:Proceedings of 4th IEEE International Workshop on Perfor mance Evaluation of Tracking and Surveillance, Graz, Austria,2003:24-31.
[10]Y Wu, T S Huang. Robust visual tracking by integrating multiple cues based on co-inference learning[J]. International Journal of Computer Vision,2004,58(1): 55-71.
[11]曹银花,李林,郜广军,等.动摄像机和动目标跟踪模式下的目标检测新方法[J].光学技术,2005,31(2)：276-278.
[12]邵文坤,黄爱民,韦庆.动态场景下的运动目标跟踪方法研究[J].计算机仿真,2006,23(5)：181-184.
[13]郑世民.动态场景图像序列中运动目标检测与跟踪：[博士学位论文].南京：东南大学,2005.1-12.
[14]S Lampotang, D E, Lizdas J, et al. Panoramic screen-based simulation with dynamic background. Simulation Conference[C], WSC2008. Winter:2947-2947.
[15]R Collins, A Lipton. A system for video surveillance and monitoring:VSAM final report. CMU-RI-TR-00-12[J]. Robotic Institute Carnegie Mellon University. 2000.
[16]王亮,胡卫明,谭铁牛.人运动的视觉分析综述[J].计算机学报.2002(3)：1-16.
[17]V Mahadevan, N Vasconcelos. Background subtraction in highly dynamic sce-nes. Computer Vision and Pattern Recognition[C], IEEE Conference on CVPR. 2008:1-6.
[18]I Visentini. On Stable Dynamic Background Generation Technique Using Gaussian Mixture Models for Robust Object Detection. Advanced Video and Signal Based Surveillance[C], IEEE International Conference on AVSS'08.2008: 41-48.
[19]L Liu, Y J Wang. Robotic Dynamic Target Recognition and Tracking Based on the Monoeular Vision[C]. Control Conference, Chinese.2008:193-197
[20]Lee J S, Rhee K Y, Kim S D. Moving Target Tracking Algorithm Based on the Confidence Measure of Motion Vectors[C]. Proc of the IEEE In the Conf on Image Processing,2001:369-372.
[21]Thakoor N, Gao J. Automatic Video Object Shape Extraction and Its Classifica-tion with Camera in Motion[C]. Proccedings of IEEE In the Conf on Image Processing,2005:437-440.
[22]郑世友,费树岷.动态场景图像序列中运动目标检测新方法[C].中国图象图形学报2007(12)：1590-1597.
[23]Z Chen, S T Birchfield. Person Following with a MobileRobot Using Binocular Feature-Based Tracking[J]. IROS,2007:815-820.
[24]Wang L, Hu W M, Tan T N. A survey of visual analysis of human motion[J]. Chinese Journal of Computers,2002,25(3):225-237.
[25]Jepson A, Fleet D, EI-Maraghi T. Robust online appearance models for visual tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003,25(10):1296-1311.
[26]Matthews I, Ishikawa T, Baker S. The template update problem[J]. IEEE Transa-ctionson Pattern Analysis and Machine Intelligence,2004,26(6):810-815.
[27]Luo H, Eleftheriadis. A Model-based segmentation and tracking of head-and-shouder video objects for real time multimedia services[J]. IEEE Transactions on Multimedia,2003,5(3):379-389.
[28]Kato J, Watanabe T, Joga S, et al. An HMM/MRF-based stochastic framework for robust vehicle tracking[J]. IEEE Transactions on Intelligent Transportation Sys-terns.2004,5(3):142-154.
[29]Djuric P M, Kotecha J H, ZhangJ, et al. Particle filtering[J]. IEEE Signal Process-ing Magazine,2003,20(5):19-38.
[30]Gastand M, Barlaud M, Aubert G Combining shape Prior and statistical features for active contour segnlentation[J]. IEEE Transactions on Circuits and Systems for Video Technology,2004,14(5):726-734.
[31]Kap-hos, Ju-jang L. Object tracking using adaptive color Snake model[C]. in: Proeeedings 2003 IEEE/ASME International Conference on Advanced Intelligent Mechattronies(AIM2003),2003.1406-1410.
[32]Kim W, Lee J. Visual tracking using snake based on targets contour information [C]. in:IEEE International SymPosium on Industrial Eleetronies. Pusan. Institute of Eleetrieal and Electronies Engineers Inc,2001:176-180.
[33]V Mahadevan, N Vasconcelos. Background subtraction in highly dynamic seen-es[C]. Computer Vision and Pattern Recognition, CVPR 2008. IEEE Conference on,2008:1-6.
[34]S Lampotang, D E Lizdas, J J Tumino. Panoramic screen-based simulation with dynamic background[C]. Simulation Conference, Winter,2008:2937-2947.
[35]A J Lipton, H Fujiyoshi, R S Patil. Moving Target Classification and Tracking from Real-time Video[C], in Proc. of the 4th IEEE Workshop on Application of Computer Vision,1998:8-14.
[36]M Piaggio, P Fornaro, A Piombo, et al. An optical-flow person following behave-iour[C], In IEEE ISIC/CIRNISAS Joint Conference,1998:4078-4083.
[37]G Chivil'o, F Mezzaro, A Sgorbissa, et al, Follow the leader behaviour through optical flow minimization[C], IEEE International Conference on Intelligent Robots and Systems IROS,2004(4):3182-3187.
[38]A Handa, J Sivaswamy, K M Krishna, et al. Person Following with a Mobile Robot Using a Modified Optical Flow[J], Advances in Mobile Robotics, World Scientific,2008:1154-1160
[39]Heuer J, Kaup A. Global motion estimation in image sequences using robust motion vector field segmentation[C], In:Proceeding of the 7th ACM internation-al conference on Multimedia, Sydney, Australia,1999:261-264.
[40]Cheung H K, Siu W C. Fast global motion estimation for sprite generation[J], In: Proceeding of the IEEE International Symposium on Circuits and Systems (ISCAS) Arizona, USA,2002(3):26-29.
[41]Sorwar G, Mursher M, Dooley L. Fast global motion estimation using iterative least-square technique[C], In:Proceedings of the 4th International Conference on Information Communications&Signal, Proceedings of the 4th IEEE Pacific-Rim Conference on Multimedia, Singapore,2003(1):282-286.
[42]D G Lowe. Distinctive Image Features from Scale Invariant Keypoints[J], Intern-ational Journal of Computer Vision,2004,60(2):91-111.
[43]D G Lowe. Local Feature View Clustering for 3D Object Recognition[J]. IEEE Conference on Computer Vision and Pattern Recognition, Kauai, Hawaii,2001(1): 682-688.
[44]T Lindberg. Scale-Space for discrete Signals[J]. IEEE Trans. PAMI,1990(12): 234-254.
[45]M Brown, D G Lowe, Invariant Features from Interest Point Groups[C]. British Machine Vision Conference, BMVC 2002, Cardiff,Wales:British Machine Vision Association,2002:656-665.
[46]F Crow. Summed-area tables for texture mapping[J]. Proceedings of SIGGRAPH, 1984,18(3):207-212.
[47]P Viola, M Jones. Rapid object detection using a boosted cascade of simple features[J]. CVPR,2001(1):511-518.
[48]Hartley R Zisserman. A Multiple view geometry in computer vision[M]. Cambri-dge University Press 2000.
[49]K Inui, S Kaneko, S Igarashi. Robust line fitting using LmedS clustering, Syste-ms and Computers in Japan[J].2003,34:92-100.
[50]Fischler M, Bolles R. Random sample consensus:A paradigm for model fitting with applications to image analysis and automated cartography[J], Communi-caions of the ACM,1981(24):381-385. -caions of the ACM,1981(24):381-385.
[51]S P Lloyd. Least Squares Quantization in PCM, IEEE Transactions on Inform-ation Theory 28[J].1982:129-137.
[52]Keinosuke Fukunage, Larry D Hosterler. The estimation of the gradient of a density function, with applications in pattern recognition[J]. IEEE Transactions oninformation theory, January 1975(1):5-20.
[53]Yizong Cheng. Mean Shift, mode seeking and clustering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1995:8-17.
[54]Comaniciu D. Nonparametric information fusion for motion estimation[C]. In: Proc.Of the IEEE Conf.on Computer Vision and Pattern Recognition (CVPR). 2003:59-66.
[55]Comaniciu D, Ramesh V, Bue AD.Multivariate saddle point detection for statistic-cal clustering[C]. In:Proc.of the European Confence on Computer Vision (ECCV).2002:23-26.
[56]A Bhattacharyya. On a measure of divergence between two statistical populations defined by their probability distributions[J]. Bull.Calcutta Math,1943:99-109.
[57]左军毅,梁彦,赵春晖.Mean Shift跟踪算法中尺度自适应策略的研究[J].中国图象图形学报,2008,13(9)：1750-1757.
[58]Yilmaz A. Object Tracking by Asymmetric Kernel Mean Shift will Automatic Scale and Orientation Selection[C]. IEEE Conf. On Computer Vision and Pattern Recognition,2007:1-6.
[59]COMANICIU D, RAMESH V, MEER P. Kernel-based object tracking[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(5):564-577.
[60]R E Kalman, A new approach to linear filtering and prediction problems[J], Journal of Basic Engineering 1960:35-45.
[61]Gregory D Halter, Peter N Belbumeur.Efficient Region Tracking With Parametric Models of Geometry and Illumination[J]. IEEE Transactions On Pattern Analysis And Machine Intelligence, October 1998(10):1025-1038.
[62]S Th run, W Burgard, and D Fox, Probabilistic Robotics:Intelligent Robotics and Autonomus Agents[M], Cambridge, MA:MIT Press,2005.