动态图像序列中运动目标检测若干技术问题的研究
|
摘要
目标检测与跟踪技术是计算机视觉的主要研究方向之一,它是智能监控、人机交互、移动机器人视觉导航、全景战车等应用的基础和关键技术。动态图像序列中,除了具有与图像一样的空间特性外(如颜色信息、纹理信息等),还具有时序特性,即动态图像序列中的运动信息。因此,与传统的图像分析技术相比,基于动态图像序列的运动信息分析在上述领域中起着更加重要的作用,然而在实际的应用中,由于摄像机抖动、运动目标的不规则运动以及摄像机运动所引起的背景运动等因素,使得运动目标的检测变得困难,本文对于动态背景下图像序列中的运动目标检测展开了研究,研究内容和主要成果如下:
(1)针对动态背景下微小运动目标的检测,提出了基于帧间差分向量一阶范数检测技术与空域峰值点双窗口搜索技术相结合的检测方法,该方法不仅能检测出帧间位移不小于一个像元的点目标,而且能检测出多帧累积位移大于一个像元的点目标,实验结果表明,该方法能有效地检测出动态背景下的多个运动点目标。
(2)提出了基于六参数仿射模型的全局运动估计补偿检测法,全局运动估计是指对图像序列中造成背景运动的摄像机运动进行估计。通过全局运动估计,可以获得帧间像素的相关性,用得到的全局运动估计参数对图像序列进行补偿,从而将复杂运动背景下的运动分析转化为静态背景下的运动。在本论文中,该算法对动态序列图像中的运动目标(如运动车辆、行人等)检测有不错的效果。
(3)对上面全局运动估计中使用的宏块匹配算法进行了优化,从而降低了计算量,提高了程序的运行速度,增加了检测的可靠性。
The technology of object detection and tracking is one of the hotspots in the field of computer vision, which is also the basic and important technology in the application of smart surveillance, human-machine interface, mobile Robert navigation, full-sight caterpillar system and so on. Dynamic image sequences possess not only spatial properties like images (such as color, texture etc), but also temporal properties, namely, the motion information. Thus, compared with the traditional image analysis techniques, motion information analysis in image sequences plays more important role in solving the above mentioned problems. However, in the practical applications, jittering camera, moving objects’irregular motion, the moving background caused by camera motion etc make the detection of moving objects difficult. A thorough research on the detection of moving objects in dynamic image sequences has been carried through in this dissertation. Main research works and achievements of this dissertation are listed as follows:
(1) A method of detecting moving point objects in dynamic image sequences is proposed, which is based on the combination of two methods: detecting method based on first-order norm of frame difference vectors and the method based on peak value detection in airspace searched by double windows. It can detect not only the point object whose displacement is no less than one pixel between two continuous frames, but also the point object whose displacement is no more than one pixel between two continuous frames and whose displacement gathered of multiple continuous frames is larger than one pixel. The result indicates that, this method provides a new solution to detect multiple moving point objects in dynamic image sequences effectively.
(2) A method is proposed to estimate global motion, which is based on 6-parameter affine model. The method is to estimate the law of the camera motion, which causes the background moving in the image sequences. By the estimation of global motion, the pixel correspondences between adjacent frames can be attained, and frames can be compensated with the parameters get by the global motion estimation, thus the problem with complex moving background can be simplified to the one with static background. This method has a fine effect in the detection of moving objects (such as moving cars, moving people etc) in dynamic image sequences.
(3) The block matching algorithm used in the global motion estimation is optimized, which reduces the calculate amount, increase the speed of program running, and increase the reliability of moving objects detection to a great extent.
(1)针对动态背景下微小运动目标的检测,提出了基于帧间差分向量一阶范数检测技术与空域峰值点双窗口搜索技术相结合的检测方法,该方法不仅能检测出帧间位移不小于一个像元的点目标,而且能检测出多帧累积位移大于一个像元的点目标,实验结果表明,该方法能有效地检测出动态背景下的多个运动点目标。
(2)提出了基于六参数仿射模型的全局运动估计补偿检测法,全局运动估计是指对图像序列中造成背景运动的摄像机运动进行估计。通过全局运动估计,可以获得帧间像素的相关性,用得到的全局运动估计参数对图像序列进行补偿,从而将复杂运动背景下的运动分析转化为静态背景下的运动。在本论文中,该算法对动态序列图像中的运动目标(如运动车辆、行人等)检测有不错的效果。
(3)对上面全局运动估计中使用的宏块匹配算法进行了优化,从而降低了计算量,提高了程序的运行速度,增加了检测的可靠性。
The technology of object detection and tracking is one of the hotspots in the field of computer vision, which is also the basic and important technology in the application of smart surveillance, human-machine interface, mobile Robert navigation, full-sight caterpillar system and so on. Dynamic image sequences possess not only spatial properties like images (such as color, texture etc), but also temporal properties, namely, the motion information. Thus, compared with the traditional image analysis techniques, motion information analysis in image sequences plays more important role in solving the above mentioned problems. However, in the practical applications, jittering camera, moving objects’irregular motion, the moving background caused by camera motion etc make the detection of moving objects difficult. A thorough research on the detection of moving objects in dynamic image sequences has been carried through in this dissertation. Main research works and achievements of this dissertation are listed as follows:
(1) A method of detecting moving point objects in dynamic image sequences is proposed, which is based on the combination of two methods: detecting method based on first-order norm of frame difference vectors and the method based on peak value detection in airspace searched by double windows. It can detect not only the point object whose displacement is no less than one pixel between two continuous frames, but also the point object whose displacement is no more than one pixel between two continuous frames and whose displacement gathered of multiple continuous frames is larger than one pixel. The result indicates that, this method provides a new solution to detect multiple moving point objects in dynamic image sequences effectively.
(2) A method is proposed to estimate global motion, which is based on 6-parameter affine model. The method is to estimate the law of the camera motion, which causes the background moving in the image sequences. By the estimation of global motion, the pixel correspondences between adjacent frames can be attained, and frames can be compensated with the parameters get by the global motion estimation, thus the problem with complex moving background can be simplified to the one with static background. This method has a fine effect in the detection of moving objects (such as moving cars, moving people etc) in dynamic image sequences.
(3) The block matching algorithm used in the global motion estimation is optimized, which reduces the calculate amount, increase the speed of program running, and increase the reliability of moving objects detection to a great extent.
引文
[1]周兵,运动对象检测及其在视频监控中的应用:[博士学位论文],北京;北京航空航天大学,2003
[2]吴江华,运动目标自动检测与跟踪方法:[硕士学位论文],哈尔滨;哈尔滨工业大学,1999
[3]Beymer D,McLauchlan P.F,Coifman B,and Malik,A real-time computer vision system for measuring traffic parameters,In Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition,1997
[4]Beymer D,Konolige K,Real-time tracking of multiple people using continuous detection,In Proc. International Conference on Computer Vision,1999
[5]Collins R.T,Lipton A.J,Kanade T,A system for Video Surveillance and Monitoring,American Nuclear Society Eight Internet Optical Meeting on Robotics and Remote Systems,1999
[6]Mataric M.J,Learning in Multi-Robot Systems,Adaptation and Learning in Multi-Agent Systems,Lecture Notes in Artificial Intelligence,1998,152~163
[7]Barron J,Fleet D,Beauchemin S,Performance of optical flow techniques, International Journal of Computer Vision,1994,12(1),42~77
[8]Ketani A,Kuno Y,Shimada N,et al,Real time Surveillance System Detecting Persons in Complex Scenes,Proceedings of Image Analysis and Processing,1999,1112~1115
[9]Wildes R,Wixson L,Detecting Salient Motion Using Spatintemporal Filters and Optical Flow,Proc. DARPA Image Understanding Workshop,1998
[10]吉书鹏,张桂林,丁晓青,地面复杂场景图像相关跟踪算法研究,激光与红外,2002, 32(6),428~430
[11]王艳萍,实时视频图像相关跟踪的算法改进与实现,舰船科学技术,2004, 26 (3),57~60
[12]江和平,王平,沈振康,任意指定地面复杂场景图像目标的去均值相关跟踪算法,信号处理,2006, 22(1),24~27
[13]许磊,王汇源,一种改进的运动估值块匹配准则,电子技术应用,2006,32(8),35~37
[14]Hue. C,Le Cadre. J.–P,Perez. P,Sequential Monte Carlo methods for multiple target tracking and data fusion,Signal Processing, IEEE Transactions on [see also Acoustics, Speech, and Signal Processing, IEEE Transactions on],2002,50(2),309~325
[15]Bruno. M.G.S,Pavlov. A,Improved sequential Monte Carlo filtering for ballistic target tracking,Aerospace and Electronic Systems, IEEE Transactions on,2005,41(3),1103~1108
[16]Deng. L,Lee. L. J,Attias. H,Adaptive Kalman Filtering and Smoothing for Tracking Vocal Tract Resonances Using a Continuous-Valued Hidden Dynamic Model,Audio, Speech and Language Processing, IEEE Transactions on [see also Speech and Audio Processing, IEEE Transactions on],2007,15(1),13~23
[17]Deergha Rao. K,Dhawas. J. A,Parallel implementation of radar tracking extended Kalman filters on transputer networks,Aerospace and Electronic Systems, IEEE Transactions on,1995,31(2),857~862
[18]朱虹等,数字图像处理基础,北京:科学出版社,2005,4
[19]姚敏等,数字图像处理,北京:机械工业出版社,2006,1
[20]周长发,精通Visual C++图像处理编程(第三版),北京:电子工业出版社,2006,6
[21]阮秋琦,阮宇智等(译),数字图像处理(第二版),北京:电子工业出版社,2004,6
[22]杨淑莹,VC++图像处理程序,北京:清华大学出版社,2003,11
[23]R. L. Joshi,H. Jafarkhani,et al,Comparison of different methods of classification in subband coding of images [J] ,IEEE Transactions on Image Processing,1997,6(11),1473~1486
[24]J. Wang,K. Zhang,S. Tang,Spectral and spatial decorrelation of Landsat-TM data for lossless compression [J] ,IEEE Transactions on Geoscience and Remote Sensing,1995,33(5),1277~1285
[25]伏思华,张小虎,基于序列图像的运动目标实时检测方法,光学技术,2004,30(2),215~217
[26]高文,陈熙霖,计算机视觉——算法与系统原理,北京:清华大学出版社,广西:广西科学技术出版社,2000,83~100
[27]章毓晋,图像工程——图像理解与计算机视觉,北京:清华大学出版社,2000,119~126
[28]张会军,复杂背景下的动目标检测技术,电光系统,2004,1(107),19~23
[29]B.K.P. Horn,B.G. Schunck,Determining Optical flow,Artificial Intelligence,August,1981,185~203
[30]Minami YAMADA, et al., Estimation of Detecting Reliability and TrackingCondition for a Picture Tracking System of Moving Targets, Proc. CVPR, 1983
[31]黎西龙,末制导成像跟踪方案设计:[硕士学位论文],西安:西安电子科技大学,1988
[32]顾樑，含背景运动的图像序列中运动目标跟踪技术的研究：[硕士学位论文],上海;上海交通大学,2003
[33]蔡晓钧,基于图像序列的运动小目标检测:[硕士学位论文],长沙;国防科学技术大学,2003
[34]陈颖,序列图像中微弱点状运动目标检测及跟踪技术研究:[博士学位论文],西安;电子科技大学,2002
[35]屈有山,田维坚,李英才等,基于隔帧差分向量无穷范数的运动点目标检测,红外与激光工程,2003,32(2),157~162
[36]张飞,李承芳,史丽娜等,基于隔帧差分向量无穷范数的运动弱小目标的检测,激光与红外,2005,35(7),531~533
[37]Heuer J.,Kaup A.,Global motion estimation in image sequences using robust motion vector field segmentation,In:Proceeding of the 7th ACM international conference on Multimedia,Sydney,Australia,1999,261~264
[38]He Y.W.,Feng B.,Yang S,Q,,Zhong Y.C.,Fast global motion estimation for global motion compensation coding,In:Proceedings of the IEEE International Symposium on Circuits and Systems(ISCAS),2001,2,233~236
[39]Cheung H.K.,Siu W.C.,Fast global motion estimation for sprite generation,In:Proceeding of the IEEE International Symposium on Circuits and Systems(ISCAS),Arizona,USA,2002,3,26~29
[40]Paik J.K.,Park Y.C.,An edge detection approach to digital image stabilization based on tri-state adaptive linear neurons,IEEE Transactions on Consumer Electronics,1991,37(3),521~530
[41]Sorwar G.,Mursher M.,Dooley L,Fast global motion estimation using iterative least-square technique,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]Ko S.,Lee S.,Kang E.,Fast digital image stabilizer based on gray-coded bit-plane matching,IEEE Transactions on Consumer Electronics,1999,45(3):598~603
[43]王嘉,王海峰,刘青山等,基于三参数模型的快速全局运动估计,计算机学报,2006,29(6),920~927
[44]吴思,视频运动信息分析技术研究:[博士学位论文],沈阳:中国科学院计算技术研究所,2005
[45]Hoetter M.,Differential estimation of the global motion parameters zoom and pan,Signal Processing,1999,16,249~265
[46]Zervakis M.E.,Venetsanopoulos A.N.,Iterative least squares estimation in nonlinear image restoration,IEEE Transactions on Signal Processing,1992,4(4),927~945
[47]陈敏慎,运动目标物体实时追踪图像匹配法的研究:[硕士学位论文],哈尔滨:哈尔滨工业大学,2003
[48]Li W.,Salari E.,Successive elimination algorithm for motion estimation,IEEE Transactions on Image Processing,1995,4(1),105~107
[2]吴江华,运动目标自动检测与跟踪方法:[硕士学位论文],哈尔滨;哈尔滨工业大学,1999
[3]Beymer D,McLauchlan P.F,Coifman B,and Malik,A real-time computer vision system for measuring traffic parameters,In Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition,1997
[4]Beymer D,Konolige K,Real-time tracking of multiple people using continuous detection,In Proc. International Conference on Computer Vision,1999
[5]Collins R.T,Lipton A.J,Kanade T,A system for Video Surveillance and Monitoring,American Nuclear Society Eight Internet Optical Meeting on Robotics and Remote Systems,1999
[6]Mataric M.J,Learning in Multi-Robot Systems,Adaptation and Learning in Multi-Agent Systems,Lecture Notes in Artificial Intelligence,1998,152~163
[7]Barron J,Fleet D,Beauchemin S,Performance of optical flow techniques, International Journal of Computer Vision,1994,12(1),42~77
[8]Ketani A,Kuno Y,Shimada N,et al,Real time Surveillance System Detecting Persons in Complex Scenes,Proceedings of Image Analysis and Processing,1999,1112~1115
[9]Wildes R,Wixson L,Detecting Salient Motion Using Spatintemporal Filters and Optical Flow,Proc. DARPA Image Understanding Workshop,1998
[10]吉书鹏,张桂林,丁晓青,地面复杂场景图像相关跟踪算法研究,激光与红外,2002, 32(6),428~430
[11]王艳萍,实时视频图像相关跟踪的算法改进与实现,舰船科学技术,2004, 26 (3),57~60
[12]江和平,王平,沈振康,任意指定地面复杂场景图像目标的去均值相关跟踪算法,信号处理,2006, 22(1),24~27
[13]许磊,王汇源,一种改进的运动估值块匹配准则,电子技术应用,2006,32(8),35~37
[14]Hue. C,Le Cadre. J.–P,Perez. P,Sequential Monte Carlo methods for multiple target tracking and data fusion,Signal Processing, IEEE Transactions on [see also Acoustics, Speech, and Signal Processing, IEEE Transactions on],2002,50(2),309~325
[15]Bruno. M.G.S,Pavlov. A,Improved sequential Monte Carlo filtering for ballistic target tracking,Aerospace and Electronic Systems, IEEE Transactions on,2005,41(3),1103~1108
[16]Deng. L,Lee. L. J,Attias. H,Adaptive Kalman Filtering and Smoothing for Tracking Vocal Tract Resonances Using a Continuous-Valued Hidden Dynamic Model,Audio, Speech and Language Processing, IEEE Transactions on [see also Speech and Audio Processing, IEEE Transactions on],2007,15(1),13~23
[17]Deergha Rao. K,Dhawas. J. A,Parallel implementation of radar tracking extended Kalman filters on transputer networks,Aerospace and Electronic Systems, IEEE Transactions on,1995,31(2),857~862
[18]朱虹等,数字图像处理基础,北京:科学出版社,2005,4
[19]姚敏等,数字图像处理,北京:机械工业出版社,2006,1
[20]周长发,精通Visual C++图像处理编程(第三版),北京:电子工业出版社,2006,6
[21]阮秋琦,阮宇智等(译),数字图像处理(第二版),北京:电子工业出版社,2004,6
[22]杨淑莹,VC++图像处理程序,北京:清华大学出版社,2003,11
[23]R. L. Joshi,H. Jafarkhani,et al,Comparison of different methods of classification in subband coding of images [J] ,IEEE Transactions on Image Processing,1997,6(11),1473~1486
[24]J. Wang,K. Zhang,S. Tang,Spectral and spatial decorrelation of Landsat-TM data for lossless compression [J] ,IEEE Transactions on Geoscience and Remote Sensing,1995,33(5),1277~1285
[25]伏思华,张小虎,基于序列图像的运动目标实时检测方法,光学技术,2004,30(2),215~217
[26]高文,陈熙霖,计算机视觉——算法与系统原理,北京:清华大学出版社,广西:广西科学技术出版社,2000,83~100
[27]章毓晋,图像工程——图像理解与计算机视觉,北京:清华大学出版社,2000,119~126
[28]张会军,复杂背景下的动目标检测技术,电光系统,2004,1(107),19~23
[29]B.K.P. Horn,B.G. Schunck,Determining Optical flow,Artificial Intelligence,August,1981,185~203
[30]Minami YAMADA, et al., Estimation of Detecting Reliability and TrackingCondition for a Picture Tracking System of Moving Targets, Proc. CVPR, 1983
[31]黎西龙,末制导成像跟踪方案设计:[硕士学位论文],西安:西安电子科技大学,1988
[32]顾樑，含背景运动的图像序列中运动目标跟踪技术的研究：[硕士学位论文],上海;上海交通大学,2003
[33]蔡晓钧,基于图像序列的运动小目标检测:[硕士学位论文],长沙;国防科学技术大学,2003
[34]陈颖,序列图像中微弱点状运动目标检测及跟踪技术研究:[博士学位论文],西安;电子科技大学,2002
[35]屈有山,田维坚,李英才等,基于隔帧差分向量无穷范数的运动点目标检测,红外与激光工程,2003,32(2),157~162
[36]张飞,李承芳,史丽娜等,基于隔帧差分向量无穷范数的运动弱小目标的检测,激光与红外,2005,35(7),531~533
[37]Heuer J.,Kaup A.,Global motion estimation in image sequences using robust motion vector field segmentation,In:Proceeding of the 7th ACM international conference on Multimedia,Sydney,Australia,1999,261~264
[38]He Y.W.,Feng B.,Yang S,Q,,Zhong Y.C.,Fast global motion estimation for global motion compensation coding,In:Proceedings of the IEEE International Symposium on Circuits and Systems(ISCAS),2001,2,233~236
[39]Cheung H.K.,Siu W.C.,Fast global motion estimation for sprite generation,In:Proceeding of the IEEE International Symposium on Circuits and Systems(ISCAS),Arizona,USA,2002,3,26~29
[40]Paik J.K.,Park Y.C.,An edge detection approach to digital image stabilization based on tri-state adaptive linear neurons,IEEE Transactions on Consumer Electronics,1991,37(3),521~530
[41]Sorwar G.,Mursher M.,Dooley L,Fast global motion estimation using iterative least-square technique,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]Ko S.,Lee S.,Kang E.,Fast digital image stabilizer based on gray-coded bit-plane matching,IEEE Transactions on Consumer Electronics,1999,45(3):598~603
[43]王嘉,王海峰,刘青山等,基于三参数模型的快速全局运动估计,计算机学报,2006,29(6),920~927
[44]吴思,视频运动信息分析技术研究:[博士学位论文],沈阳:中国科学院计算技术研究所,2005
[45]Hoetter M.,Differential estimation of the global motion parameters zoom and pan,Signal Processing,1999,16,249~265
[46]Zervakis M.E.,Venetsanopoulos A.N.,Iterative least squares estimation in nonlinear image restoration,IEEE Transactions on Signal Processing,1992,4(4),927~945
[47]陈敏慎,运动目标物体实时追踪图像匹配法的研究:[硕士学位论文],哈尔滨:哈尔滨工业大学,2003
[48]Li W.,Salari E.,Successive elimination algorithm for motion estimation,IEEE Transactions on Image Processing,1995,4(1),105~107