摘要
随着世界各国经济的快速发展,车辆的数量急剧增加,由此引发的交通状况问题日益受到人们的重视。智能交通系统就是为了解决这个问题而诞生的,目前已经成为全球各国政府和有关部门高度重视的高科技新领域。基于视频的闯红灯车辆自动检测系统是智能交通系统中的重要组成部分,其中有效的检测和实时跟踪是车辆的行为分析和行为判断的前提。
鉴于以上背景,本文对闯红灯车辆检测与跟踪算法进行研究。在分析和总结现有算法的基础上,提出了两个改进算法。
本文针对传统的背景提取方法在前景运动对象密度较高,亦或前景出现过于频繁的情况下,提取到的背景图像中会掺杂很多前景成分的问题,提出了一种基于形态学和像素灰度归类的背景重建算法。该算法通过三帧差分法和形态学把每帧图像区分为前景对象和背景区域。然后在用像素灰度归类方背景重建时不考虑前景对象,提高了背景出现的频率,从而能正确重建背景。通过实验证明,新方法能够在车辆密度较大的情况下能正确的生成背景。
本文为了满足车辆跟踪实时的要求,设计了一种自动初始化窗口的Mean Shift算法。另外针对Mean Shift算法在目标的颜色分布和背景相似情况下,会丢失目标的问题,利用kalman滤波器根据前面的目标位置信息来预测在本帧图像中目标的可能位置,然后用Mean Shift算法在这个位置的邻域内找到目标的真实位置。这样,我们利用了Mean Shift所没有利用的前面一帧目标的运动信息,丰富了对已知信息的使用,增强了跟踪效果。通过实验证明此算法是有效性。
With the rapid development of the world economy, the number of vehicles increase dramatically to lead to the traffic problem that is attented. Intelligent transportation system is born for solving this problem and now the global governments and relevant departments pay close attention to this new high-tech area. Automatic detection system of red light running vehicles with Video is an important component of the intelligent transportation system, in which an effective detection and real-time tracking are determine premise of behavior analysis and behavior judgment in vehicles.
View as the above background, this paper studies detection and tracking of red light running vehicle algorithm. In basis of anglicizing and summarizing existing algorithms, it proposes two improved algorithms.
Against the problem that the tradition background extraction method could not precise extraction background when moving objects have high density or appear too frequently. In this paper presented an algorithm of background reconstruction based on morphological and PIC (Pixel intensity classification). The algorithm classify the pixels in each frame into background area and moving objection area using three frame difference and Morphological analysis. Then extracting background modeling with PIC except the moving objection area, raised the frequency of background, so the background could be rebuilded proper. The experiment results show that the method proposed in this paper could extract accurate background in high vehicle density.
In order to meet the requirements of real-time vehicle tracking, it designs an Mean Shift algorithm of automatic initialization window. Another in the cases of the similar of target color distribution and background, it considers the Mean Shift Algorithm will lose target. So, according to position information in front, kalman filter is used to predict the possible position of target image in the frame, and then at the location of the neighborhood you can use the Mean Shift to find the true target location. It enriches the use of known information and enhances the tracking results by the use of target movement information of before frame. Experiment verifies the effectiveness of the method.
引文
[1]崔宝侠,李爱军.ITS常用车辆检测技术的研究[C].2005年海峡两岸智能交通运输系统学术研讨会暨第二届同舟交通论坛,2005.
[2]李晖,贾辉然,牛晓辉,宋士普.智能交通系统的研究现状及发展趋势[J].产业与科技论坛,2008,8(9):166-168.
[3]张晖,董育宁.基于视频的车辆检测算法综述[J].南京邮电大学学报,2007,3(27):88-94.
[4]王圣男,郁梅,蒋刚毅.智能交通系统中基于视频图像处理的车辆检测与跟踪方法综述[J].计算机应用.2005,9:9-14.
[5] Massimo Piccardi. Background subtraction techniques: a review [C]. IEEE International Conference on Systems, Man and Cybernetics, 2004:3099-3104.
[6]朱明旱,罗大庸,曹倩霞.帧间差分与背景差分相融合的运动目标检测算法[J].计算机测量与控制,2005,13(3):215-217.
[7]徐以美,郭宝龙,张晋.基于像素分类的运动目标检测算法[J].计算机工程,2008,23(34):205-207.
[8] Gianluca Bailo, Massimo Bariani, Paivi Ijas, Marco Raggio. Background Estimation with Gaussian distribution for Image Segmentation [J]. IEEE International Workshop on Measurement Systems for Homeland Security, Contraband Detection and Personal Safety Orlando. 2005:2-5.
[9] C. Stauffer and W.E.L. Grimson. Adaptive background mixture models for real-time tracking [C]. Proc. IEEE CVPR, June 1999:246-252.
[10] Zunbing Sheng, Xianyu Cui. An Adaptive Learning Rate GMM for Background Extraction[C]. IEEE International Conference on Computer Science and Software Engineering, pp: 174-176, 2008.
[11]侯志强,韩崇昭.基于像素灰度归类的背景重构算法[J].软件学报,2005,9(16):1568-1576.
[12] Dou Zhao, Ding Liu, Yanxi Yang. An Improved PIC Algorithm of Background Reconstruction For Detecting Moving Object [C].IEEE Fifth International Conference on Fuzzy Systems and Knowledge Discovery, 2008:24-28.
[13] HU Weiming, TAN Tieniu, WANG Liang, etal. A survey on visua surveillanee of object motion and behaviors [J].IEEE Transaction on Systems, Man and Cybemetics-Part C:Applications and Reviews,2004,34(3):334-352.
[14] Hager G, Belhumeur P. Effieient. Region tracking with parametric models of geometry and illumination [J].IEEE Transactions on Pattem Analysis and Intelligence, 1998, 20(10):1025-1039.
[15] Jepson A D, Fleet D J. Robust online appearance models for visual tracking[J].IEEE Transactions on Pattem Analysis and Machine Intelligenee.2003,25(10):1296-1311.
[16] Paragios N, Deriche R. Geodesic active contours and level sets for the detection and trackingof moving objects [J].IEEE Transactions on Patterm Analysis and Machine Intelligenee, 2000, 22(3):266-280.
[17] Paragios,N.,Deriche,R.Geodesic active regions: A new paradigm to deal with frame partition problems in computer vision [J]. Visual Communicate Represent, 2005, 13(3):249–268.
[18] Magee,D. Tracking multiple vehicles using foreground, background and motion models[J]. Image Vis.Comput, 2004, 22(2), 143-155.
[19] Kamijo,S.,Sakauchi,M. Illumination invariant and occlusion robust vehicle tracking byspatio-temporal MRF model [C]. Proc.9th World Congress on ITS, Chicago.2002.
[20] Kim, Z., Malik, J. Fast vehicle detection with probabilistic feature grouping and its application to vehicle tracking [C].Proc. IEEE Int. Conf. on Computer Vision, 2003, 524-531.
[21] Szeliski R, Terzopoulos D.Pysieally-Based and Probalistie Modeling for Computer Vision.CA, 1991, 9(l):140-152.
[22] P.Soille.形态学图像分析原理与应用[M].北京:清华大学出版社,2008:47-50,77-79.
[23]章毓晋.图像分割[M].北京:科学出版社,2001.
[24]吴薇,刘军,李旭霞.图像分割中的阈值选取方法[J].西安工业学院学报,2002,4(22):309-313.
[25]高雁飞,范九伦.一种新的灰度图像阈值分割方法[J] .西安邮电学院学报,2002 ,1(7):1-4.
[26]姚耀文,朱海洲.点阵地图图像处理过程中阈值的选取[J].华南理工大学学报,1997,3(25):25-29.
[27]安如,冯学智,王慧麟.基于数学形态学的道路遥感影像特征提取及网络分析[J] .中国图像图形学报,2003,7(8):798-804.
[28]史久根,张旺生,鲍远律.彩色地图图像中道路信息的识别和提取[J] .微机发展,2000,1: 64-66.
[29] YE Jia-ming, BAO Yuan-lu, LIU Li-ping. Road extraction from color city map [C]. Proceeding of 2002 International Conference on Control and Automation China. In Xiamen, China. 2002.260-264.
[30] Haritaoglu, I and Harwood, D. A real time system for detecting and tracking people[C]. Proceedings of Third IEEE International Conference on Automatic Face and Gesture Recognition, 1998:222-227.
[31] Stringa,E.and Regazzoni,C.S. Real-time video-shot detection for scene surveillance applications [J]. IEEE Transactions on Image Processing, Jan.2000, 9(1):69-79.
[32] J.Baorrn, D.Fleet, S.Beauehemin. Performance of optical flow techniques [J]. Intenrational Joural of Computer Vision, 1994, 12(l):42-77.
[33] A.MuratTekalp. Digital video processing,Tsinghua University press1998.
[34]王新余,张桂林.基于光流的运动目标实时检测方法研究[J] .计算机工程与应用,2004.
[35] Lobpl, Velastn SA. Automatic congestion detection system for underground platforms[C] .Proc of 2001 Int Symp on Intell Multimedia,Video and Speech Processing.2000: 158-161.
[36] Soto A, Cipriano A. Image processing applied to real time measurement of traffic flow[C] .Proc of the Twenty-Eighth South-eastern Symposium on System Theory. 1996: 312-316.
[37] J. W. Tukey. Exploratory data analysis[M], Menlo Park, CA: Addison-Wesley Publishing Co, Inc., 1971, 1977
[38] A. Prati, I. Mikic, R. Cucchiara and M. M. Trivedi. Comparative Evaluation of Moving Shadow Detection Algorithms [J]. Workshop on Empirical Evaluation Methods in Computer Vision. 2001.
[39] D. Koller, K. Daniilidis, and H. H. Nagel. Model-based Object Tracking in Monocular Image Sequences of Road Traffic Scenes [J]. International Journal of Computer Vision. 1993: 10(3), 257-281.
[40] M. Bejanin, A. Huertas, G. Medioni, and R. Nevatia. Model Validation for Change Detection [J]. International IEEE Workshop on Applications of Computer Vision. 1994:160-167
[41] C. Jiang and M. O. Ward. Shadow Segmentation and Classification in a Constrained Environment [J]. CVGIP: Image Understanding. 1994: 59(2), 213-225.
[42]陈柏生,陈锻生.基于归一化RGB彩色模型的运动阴影检测[J] .计算机应用,2006,26(8) :1879-1881.
[43] I. Mikic, P.C. Cosman, G.T. Kogut, et al. Moving shadow scenes[C]. Proceedings of International Conference on Pattern and object detection in traffic Recognition, 2000. 321-324
[44] L.P .Liu, J. M. Xu, and H. Y Wang. A New Vehicle Shadow Handler Based on Texture Invariance [J]. Journal of Wuhan University of Technology (Transportation Science &Enginneering), 2005, 29, pp: 1005-1008.
[45] A. Leone, C. Distante, and F. Buccolieri, "A texture-based approach for shadow detection," Proceedings[C]. IEEE Conference on Advanced video and Signal Based Surveillance, 2005. pp. 371-376.
[46] R. Cucchiara, C. Grana, M. Piccardi, et al. Improving shadow suppression in moving object detection with HSV color information[C]. Proceedings of IEEE intelligent transportation Systems Conference, 2001:334-339.
[47] CHEN Baisheng, LEI Yun qi. Indoor and outdoor people detection and shadow suppression by exploiting HSV color information[C]. Proceedings of the Fourth International Conference on Computer and Information Technology, 2004, pp: 137-142.
[48]张丽,李志能.基于阴影检测的HSV空间自适应背景模型的车辆追踪检测[J] .中国图像图形学报,2004,8(7):778-782.
[49] R. Cucchiara et al. Improving Shadow Suppression in Moving Object Detection with HSV Color Information[C]. Proc of Intel. Transportation Systems Conf, 2001:334-339.
[50] Nicolas Martel-Brisson and Andre Zaccarin. Moving cast shadow detection from a Gaussian mixture shadow model[C]. Proceedings of 2005 IEEE Computer Society Conference onComputer Vision and Pattern Recognition, San Diego, CA, USA, Papers by Session.
[51]王军利,王树根.一种基于RGB彩色空间的影像阴影检测方法[J] .信息技术,2002,26(12):7-9.
[52] Rosin P L,Ellis T.Image Difference Threshold Strateties and Shadow Detection[C]. Pro.of the 6th British Machine Vision Conference, 1994:347-356.
[53]李忠武,高广珠等.图像序列目标检测中阴影的消除[J] .计算机应用研究,2000,17(12):19-20.
[54] E. Salvador, A. Cavallaro, and T. Ebrahimi. Shadow Identification and Classification using Invariant Color Models. ICASSPOI. 2001. 1545-1548.
[55] T. Gevers and A. W. M. Smeulders. Color-Based Object Recognition [J]. Pattern Recognition. 1999. 32(3):453-64.
[56] SHAFER SA. Using color to separate reflection components [J]. Color research and applications, 1985, 10(4): 210-218.
[57] Wang L, Hu W M, Tan T N. A survey of visual analysis of human motion [J]. Chjnese Journal of Computers, 2002, 25(3):225~237.
[58] Murat TekalpA.数字视频处理[M] .电子工业出版社,1998.
[59] Jorge Badenas, Bober M, and Pla F. Motion and intensity-based segmentation and its application to traffic monitoring[C]. International Conference on Image Analysis and Processings ICIAP’97, Florence, Italy.1997:502-509.
[60] Jorge Badenas and Pla F. Segmentation based on region-tracking in image sequences for traffic monitoring[C]. In 14th International conference on pattern Recognition, 1998, pages: 999-1001.
[61] Smith S, Brady J. SUSUN-A new approach to low level image processing [J]. International Journal of Computer Vision, 1997, 23(l):45~78.
[62] Liu Y, Hunag TS. Determining straight line correspondences from intensity images. Pattern Recognition, 1991, 24(6):489-504.
[63] Jung S K, Wohn K Y, A model-based 3-D tracking of rigid objects from a sequence of multiple perspective views [J]. Engineering Applications of Artificial Intelligence, 2000, 13:215-236.
[64]周志宇,汪亚明,黄文清.基于动态图像序列的运动目标跟踪[J].浙江工程学院学报, 2002,19(3):165-170.
[65] Kass M, Witkin A, Terzopoulos D. Snakes: Active contour models [J]. International Journal of Computer Vision, 1988, 1(4): 321-331.
[66] Won Kim, Choon-Young Lee, Ju-Jang Lee, Tracking moving object using Snakes jump based on image flow [J], MECHATRONICS, 2001, 11:199-216.
[67] Koller D, Daniilidis K, Nagel H. Model-based object tracking in monocular image sequencesof road traffic scene [J]. International Journal of Computer Vision, 1993, 10(3):257-281.
[68] Genneyr.D.Traeking known three dimensional objeets[C]. Proe. National Conefrence on Artificial Intelligenee, 1982: 13-17.
[69] Niekels K, Hutchinson S. Estimating uncertain in SSD-based feature tracking [J]. Imgae and Vision Computing, 2002, 20: 47-58.
[70] Cheng YZ. Mean shift, mode seeking, and clustering [J]. IEEE Trans. on Pattern Analysis and Machine Intelligence, 1995, 17(8): 790-799.
[71] Comaniciu D, Meer P. Mean shift: A robust approach toward feature space analysis [J]. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2002, 24(5):603-619.
[72] Comaniciu D, Ramesh V, Meer P. Real-Time tracking of non-rigid objects using mean shift [C]. In: Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2000:142-149.
[73] Collins RT. Mean shift blob tracking through scale space [C]. In: Proc. of the Conf. on Computer Vision and Pattern Recognition (CVPR). 2003. 18?20.
[74] Georgescu B, Shimshoni I, Meer P. Mean shift based clustering in high dimensions: A texture classification example [C]. In: Proc. of the ICCV. 2003. 456?463.
[75]刘玉秋.基于视频流的车辆检测算法[D] .吉林大学,2008-4
[76]孔繁奇.基于视频图像的车辆检测跟踪技术的分析研究[D] .大连理工大学,2007-12.
[77]陈燕兵.基于视频检测技术的高速公路超速抓拍系统[D] .浙江大学,2007-5.
[78]谢寒生,盛翊智,张绍满.基于快速匹配算法的交通监控系统[J] .微机发展,2004,2(14):23-25.
[79]赵跃,王世平,杨延西,刘丁.基于背景重建的多运动目标检测与跟踪[J] .西安理工大学学报,2007,1(23):6-9.
[80]张羽.智能交通中闯红灯车辆视频检测算法的研究[D] .大连海事大学,2006.
