摘要
随着经济的高速发展,我国汽车保有量急剧增长,很多城市出现了“停车难”问题。露天停车场的科学管理对减少道路拥堵,保证城市交通安全有重要意义,而对其进行视频监控是实现科学管理的有效手段,在这其中运动车辆的检测与跟踪是技术关键。
本文的主要研究内容是基于视频的运动车辆检测和跟踪问题,其中涉及到视频环境下的背景实时更新与生成、运动车辆的检测、运动车辆的跟踪和重叠车辆的分割等内容。
在运动车辆检测方面,本文对背景差分法进行了改进,提出一种新的背景生成和更新算法。露天停车场场景的变化频率较为缓慢,因此可以认为车辆监控视频图像中每个像素的灰度值在较长一段时间内会稳定分布在某一个确定的范围,本文就根据图像中各个像素点的灰度取值所属区间次数进行背景提取;在背景更新时,考虑当背景中含有长时间静止车辆时的背景重构问题,结合边缘检测判断是否进行背景更新,如果静止车辆的状态发生变化,则利用前景图中对应区域的场景进行背景重构。
在运动车辆跟踪方面,将跟踪过程分成特征提取、运动预测和目标匹配三部分。特征提取主要是来描述目标的形状特征(如运动目标的质心,目标外接矩形的大小);目标运动估计模块利用Kalman滤波器预测被跟踪运动目标在下一帧中可能处于的位置,确定目标搜索范围;目标匹配模块是寻找运动目标在图象序列各帧中的对应关系,从而确定车辆的运动轨迹。
车辆跟踪过程经常出现车辆遮挡重叠问题,它会使检测区域出现误检而导致跟踪目标丢失。针对以上问题,本文中首先使用运动区域的宽高比和占空比作为指标进行误检判断,然后用基于对应顶点车辆分割算法对含有多个车辆的区域进行切分。
Along with the rapid development of economy, the amount of motor vehicles has a rapid growth in China. As a result, the difficulty of parking has perplexed many cities in recent years. The scientific management of open parking lots is very significant to reduce traffic jam and insure traffic safety. Video surveillance is effective technology means, in which moving vehicle auto-detecting and auto-tracking is the basic part.
The thesis mainly discusses the detection of vehicle based on video. It involves in the following several topics, back ground update and extraction, moving vehicle detection, moving object tracking and moving vehicles segmentation.
In first part of moving object detecting and retrieving,the paper propose a new algorithm of background extraction and updating. The interval pixel value belongs to in the video stream is calculated to extract background. The paper take advantage of edge detection method to realize background updating under the case that the current background image contains stagnant vehicle by analyzing area of continuous change. Testing results demonstrate that when the method is applied,the self-adaptive updating of background can be realized with a high accuracy, meanwhile it is simple to implement and suitable for real-time detection.
For real-time tracking of moving target, this thesis present a method of tracking moving vehicle based on Kalman filtering, which accomplishes object tracking by feature extraction, motion estimation and object matching. Feature extraction involves describing the shape feature of present object such as the centriod and the size of its enclosing rectangle. Motion estimation is to predict the location of the tracked vehicle by means of Kalman filter in subsequent frames and ascertain the scope for searching the tracked object. Object matching is to establish one-to-one correspondence between moving objects over frames and the track of every moving object.
The moving vehicles often are sheltered by each other, it causes the false detection in target matching process. To overcome the above problem, this paper first considers the width/height ratio and occupancy ratio to make false detection judgment. Then a new moving target segmentation algorithm based on corresponding points is presented.
引文
[1]杨建全等.视频监控技术的发展与现状.现代电子技术,2006,236(21):84-91.
[2]Lipton A J,Fujiyoshih,Patlrs. Moving target classification and tracking from real-time video[C]//Proc of WACV’98.1998:8-14. Collns.R.Y, Liption.A.J, Kande.T.A system for video surveillance and monitoring[R]. Pittsburgh, USA:Camegie Mellon University,2000.
[3]Collns.R.Y,Liption.A.J,Kande.T. A system for video surveillance and monitoring [R].Pittsburgh,USA:Camegie Mellon University,2000.
[4]Haritaoglu.I, Harwood D, DavisL.W. Real-time surveillance of people and their activities. IEEE Trans Pattern Analysis and Machine Intelligence, 2000,22(8):809-830.
[5]McKenna,Setal. Tracking groups of people. Computer Vision and Image Understanding,2000,80(1):42-56.
[6]Karmann K,Brandt. A Moving object recognition using an adaptive background memory. In:CappelliniVed. Time-varying Image Processing and Moving Object Recognition.2.Elsevier,Amsterdam, The Netherlands,1990.
[7]Kilger M.A shadow handler in a video-based real-time traffic monitoring system. In:Proc IEEE Workshop on Application of Computer Vision,Palm Springs, CA,1992:1060-1066.
[8]Stauffer C,Grimson W.Adaptive background mixture models for real-time tracking,In:Proc IEEE Conference on Computer Vision and Pattern Recognition,Fort Collins,Colorado,1999,2:246-252.
[9]刘国锋,诸昌钤.光流的计算技术.西南交通大学学报.第32卷第6期,1997年12月.
[10]Meyer D,Denzler J,Niemann H.Model based extraction of articulated objects in image sequences for gait analysis.In:Proc IEEE International Conference on Image Processing,Santa Barbara,California 1997:78-81.
[11]Friedman N,Russell S.Image segmentation in video sequences:A probabilistic approach.In:Proc the Thirteenth Conference on Uncertainty in ArtificialIntelligence,Rhode Island,USA,1997.
[12]Mc Lachlan G,Krishnan T.The EM Algorithm and Extensions.Wiley Interscience ,1997.
[13]Stringa E.Morphological change detection algorithms for surveillance applications.In:Proc British Machine Vision Conference, Bristol, UK, 2000:401-411.
[14]Zivkovic Zoran, Korse Ben. An EM-like algorithm for color-histogram-based object tracing[A]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition [C], Washington,DC,USA,2004:1798-1803.
[15]Hager G,Belhumeur P. Effieient region tracking with parametric models of geometry and illumination[J],IEEE Transactions on Pattern Analysis and Intelligence.1998,20(10):1025-1039.
[16]Jepson A D, Fleet D J. Robust online appearance models for visual tracking[J]. IEEE Transactions on Pattern Analysis and Mchine Intelligenee.2003, 25(10):1296-1311
[17]M.Kass,A.P.Witkin,D Terzopoulos. Snake: active contour models[J].International Jouranl of Computer Vision.1998:321-331.
[18]Paragios N,Deriche R.Geodesic active contours and level sets for the detection and tracking of moving objects[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.2000,22(3):266-288.
[19]Meyer D,Denzler J,Niemann H.Model based extraction of articulated objects in image sequences for gait analysis.In:Proc IEEE International Conference on Image Processing,Santa Barbara,California 1997.78-81.
[20]贺贵明,李凌娟,贾振堂一种快速的基于对称差分的视频分割算法[J].小型微型计算机系统.2003,24(6):966-968.
[21]L.Maereenaro,M.Ferrari,L.Macrhesottl and C.S.Regazzoni. Multiple object tracking under heavy occlusions by using Kalman filter based on shape matching[A] , Proceedings of International Conference on Image Processing[C],Rechester, New Youk,2002(3):341-344.
[22]V.Gemiganani,M.Paterni,A.Benassi,M.Demi. Real-time contour tracking with anew edge detector[J]. Real-Time Imaging.2004,10(2):113-116.
[23]Shih,Ming-Yu, Tseng,Din-Chang. A wavlet-based multi resolution edge detection and tracking[J].Image and Vision Computing.2005,23(4):441-451.
[24]Shrif abd Kl-Azim, Imbaby Ismail, Howida abd El-latiff . An efficient object tracking technique using block-matching algorithm[A]. Nineteenth National Radio Science
[25]Yu Qifeng,Zhang Xiaohu,Lu Hongwei,Sun Xiangyi. Image sequence analysis system and a target-tracking algorithm with correlation and improved SUSAN Conference[C],Aleandria,2002:427-433.
[26]Hariharakshnan, Karthik, Schonfeld, Dan. Video tracking using block matching [A], IEEE International Conference on Image Processing[C], Barcelona, Spain ,2003(3):945-948.
[27]S.Kang,B.W.Hwang,S.W.Lee. Multiple people tracking based on temporal color feature[J].Pattern Recognition and Artificial Intelligence.2003,17(6):931-949.
[28]Karaulova I,Hall P,Marshall A.A hierarchical model of dynamics for tracking people with a single video camera[A].Proceedings of British Machine Vision Conferenee[C],Bristol,U.K,2000:352-361.
[29]Ju S,Black M,Yaccpb Y.Cardboard people:A parameterized model of articulated image motion[A].IEEE International Conference on Automatic Face and Gesture Recognition[C],Killington,Vermont USA,1996:38-44.
[30]Wachter S, Nagel H-H. Tacking persons in monocular image sequences[J]. Computer Vision and Image Understanding.1999,74(5):174-192.
[31]Paragios N,Deriche R.Geodesic active contours and level sets for the detection and tracking of moving objects[J].IEEE Transactions on Pattern Analysis and Machine Intelligence.2000,22(3):266-288.
[32]Peterfreund N. Robust tracking of position and velocity with Kalman snakes[J].IEEE Transactions on Pattern Analysis and Machine Intelligence. 2000, 22(6):564-569.
[33]E.Loutast,K.Diamantrast,I.Pitast. Occlusion resistant object tracking[A].IEEE International Conference on Image Processing[C],Thessaloniki,Greece,2001:65-68.
[34]Hieu T.Nguyen,Marcel Worring ,Rein van den Boomgaard. Occulusion robust adaptive template traking[A]. Proceedings of Eighth IEEE International Conference on Computer Vision[C], Vancouver, Candada,2001:678-683
[35]Shunsuke Karnijo,Yasuyuki Matsushita,Katsushi Ikeuchi,Masao Sakauchi. Occlusion robust tracking utilizing spatio-temporal mark of random field model[A]:Procedings of 15th International Conference on Pattern Recognition[C],Barcelona,Spain,2000:140-144.
[36]Ying Wu,Ting Yu,Guang Hua. Tacking appearances with occlusions[A].IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C],Madison, Wisconisn USA,2003:789-795.
[37]Min Hu,Weiming Hu,Tieniu Tan. Tracking People through occlusion[A].17th International Conference on Pattern Recognition[C], Cambridge, UK, 2004:724-727.
[38]K.Nummiaro,E.Koller-Meier,L.Van Gool. An adaptive color-based particle filter[J].Image and Vision Computing,2003,21(1):99-110.
[39]Yucel Altunbasak,A.Murat Tekalp. Occlusion-adaptive content-based mesh design and forward tracking[J]. IEEE Transactions on Image Processing, 1997 , 6(9):1270-1280.
[40]Jian-Wei Zhao,Peng Wang, Cong-Qing Liu. An object tracking algorithm based on occlusion mesh model[A]. International Conference on Machine Learning and Cybernetics[C],Beijing,China,2002:288-292.
[41]Rafael C. Gonzalez,Richard E.Woods,Steven L.Eddins著,阮秋琦译.数字图像处理(MATLAB版)[M].电子工业出版社,2008,4:276-282.
[42]陈天华.数字图像处理[M].清华大学出版社,2007,6:251-254.
[43]Anderson C,Bert P,Vander Wal G, Change detection and tracking using pyramids transformation techniques[J].Intelligent Robots and Computer Vision, 1985(57):72-78.
[44]廖斌,杨卫平,沈振康,基于多帧移位叠加的红外小目标检测方法[J].红外与激光工程,2002(31):150-153.
[45]马云龙,夏炯贤,吴永明,一种基于微分法的运动目标检测算法[J].计算技术与自动化,2000(4):51-54.
[46]T.Arch,A.Kaput and R.Mester,Statistical Change Detection in Moving Video[J].Signal Processing,Image Communication,Vol.1,No.2,October 1993.
[47]Kameda Yoshinari,Minoh Michihiko. A human motion estimation method using 3-successive video frames,Proceedings of International Conference on Virtual Systems and Multimedia,1996,135-140.
[48] Chen B S, Lei Y Q. Indoor and outdoor people detection and shadow suppression by exploiting HSV color information[C]/The Fourth International Conference on Computer and Information Technology.Washington,DD,USA:IEEE Computer Society ,2004:137-142.
[49] Wu Bingfei,Juang J hyhong,Tasi Pingsung, et al. A new vehicle detection approach in traffic jam conditions[J].Computational Intelligence in Image and Signal Proceessing,2007(3):1-6.
[50]潘薇,游志胜,王宁.基于模糊聚类和卡尔曼滤波的运动目标检测[J].计算机应用,2005,25(1):123-124.
[51] Zhao Ming, Li Na, Chen Chun. Statistical Inference for Automatic Video Segmentation [J]. Journal of Computer-aided Design and Computer Graphics,2005,15(3):318-323.[赵明,李娜,陈纯.采用统计推断的自动视频对象分割[J].计算机辅助设计与图形学学报,2005,15(3): 318-323.
[52]李强.一种新的自适应背景建立与更新算法[J].电子科技,2007(10):6-8.
[53]胡燕,嵇启春,李智杰,耿烨,来炳恒.视频监控中背景的提取和更新算法[J].计算技术与自动化,2009,28(1):142-144.
[54]向海霞,王卫星.车辆检测中的自适应背景更新算法研究[J].计算机工程与应用,2009,45(11):235-237.
[55]张江山,朱光喜.一种基于Kalman滤波的视频对象跟踪方法,中国图象图形学报,Vol.7(A),No.6 ,Jun.2002:606-609.