城市道路交通异常事件自动检测方法
|
摘要
针对城市道路交通流量大,易发生交通违法异常行为,提出了一种城市道路交通异常事件自动检测方法.该方法在进行异常事件检测时,首先对道路交通进行视频序列采集,然后提取视频中的车辆加速度、方向变化、几何位置信息.在提取这些特征值后,计算三个特征值变化率指数总和,并与设定的阈值进行比较,判断是否发生异常事件.通过对同一道路的相同监控视频片段采用不同的算法进行检测仿真实验,测试结果表明该方法得到的正确检测率DR为98.6%,误报率FAR为7%,相对于基于临界安全区域方法、基于Boosting检测方法其整体检测性能有所提高.
According to the situation that the flow of the city road traffic is large,and the traffic illegal behavior is easy to happen,an automatic detection method for the abnormal events of the city road traffic is presented.When using the method to detect abnormal events,video sequences of the road traffic are collected firstly,and the information of the vehicle acceleration,the direction of change,and geometrical position are extracted secondly.The sum of three feature values index change rate is calculated after extracting the feature values,and compared with the threshold to judge whether abnormal events happen.Different algorithms are used to do the simulation experiment for the video clip of the same road.The detection test results show that the correct detection rate DR of the method proposed is 98.6%,the false positive rate FAR is 7%,and the overall detection performance is improved compared with the critical security region method and the detection method based on Boosting.
According to the situation that the flow of the city road traffic is large,and the traffic illegal behavior is easy to happen,an automatic detection method for the abnormal events of the city road traffic is presented.When using the method to detect abnormal events,video sequences of the road traffic are collected firstly,and the information of the vehicle acceleration,the direction of change,and geometrical position are extracted secondly.The sum of three feature values index change rate is calculated after extracting the feature values,and compared with the threshold to judge whether abnormal events happen.Different algorithms are used to do the simulation experiment for the video clip of the same road.The detection test results show that the correct detection rate DR of the method proposed is 98.6%,the false positive rate FAR is 7%,and the overall detection performance is improved compared with the critical security region method and the detection method based on Boosting.
引文
[1]Coifman B,Beymer D,Mclauchlan P,et al.A Real-time Computer Vision System for Vehicle Tracking and Traffic Surveillance[J].Transportation Research Part C:Emerging Technologies,2008,6(4):271-288.
[2]Magee D R.Tracking Multiple Vehicles Using Foreground Background and Motion Models[J].Image and Vision Computing,2004,22(2):143-155.
[3]Haag M,Nagel H.Tracking of Complex Driving Manoeuvres in Traffic Image Sequences[J].Image and Vision Computing,1998,16(8):517-527.
[4]Masoud O,Papanikolopoulos N P.A Novel Method for Tracking and Counting Pedestrians in Real-time Using a Single Camera[J].IEEE Transactions on Vehicular Technology,2011,50(5):1267-1278.
[5]Tai J C,Tsang S T,Lin C P,et al.Real-time Image Tracking for Automatic Traffic Monitoring and Enforcement Applications[J].Image and Vision Computing,2004,22(6):485-501.
[6]马莹,吴楠楠.基于全球眼的地震现场应急救援交通路径分析系统研究[J].沈阳大学学报:自然科学版,2014,26(5):376-381.(Ma Ying,Wu Nannan.Earthquake Site Emergency Rescue Transportation Path Analysis System Based on Global Eye[J].Journal of Shenyang University:Natural Science,2014,26(5):376-381.)
[7]Pai C J,Tyan H R,Liang Y M,et al.Pedestrian Detection and Tracking at Crossroads[J].Pattern Recognition,2004,37(5):1025-1034.
[8]王自上.基于视频图像的交通事件自动识别算法研究[D].北京:北京交通大学,2010.(Wang Zishang.Automatic Recognition Algorithm Based on Video Images of Traffic Incidents[D].Beijing:Beijing Jiaotong University,2010.)
[9]刘晓男,李勃,陈启美,等.一种基于视觉技术的交通异常检测算法[J].仪器仪表学报,2011,32(S6):76-82.(Liu Xiaonan,Li Bo,Chen Qimei,et al.Vision Based Traffic Anomaly Detection Algorithm[J].Chinese Journal of Scientific Instrument,2011,32(S6):76-82.)
[10]邵士雨.基于视频的交通事件检测算法研究[D].济南:山东大学,2013.(Shao Shiyu.Research on Traffic Incident Detection Algorithm Based on Video[D].Jinan:Shandong University,2013.)
[11]Ki Y K,Lee D Y.A Traffic Accident Recording and Reporting Model at Intersections[J].IEEE Transactions on Intelligent Transportation System,2007,8(2):188-194.
[12]闻帆.基于视觉的交通路口车辆智能检测技术研究[D].哈尔滨:哈尔滨工业大学,2010.(Wen Fan.Research on Traffic Intelligent Detection Technology Based on Vision[D].Harbin:Harbin Institute of Technology,2010.)
[13]孙熙,李夏苗.基于boosting算法的交通事件检测[J].交通运输系统工程与信息,2007,7(5):37-41.(Sun Xi,Li Xiamiao.Traffic Incidents Detection Based on Boosting Method[J].Journal of Transportation Systems Engineering and Information Technology,2007,7(5):37-41.)
[2]Magee D R.Tracking Multiple Vehicles Using Foreground Background and Motion Models[J].Image and Vision Computing,2004,22(2):143-155.
[3]Haag M,Nagel H.Tracking of Complex Driving Manoeuvres in Traffic Image Sequences[J].Image and Vision Computing,1998,16(8):517-527.
[4]Masoud O,Papanikolopoulos N P.A Novel Method for Tracking and Counting Pedestrians in Real-time Using a Single Camera[J].IEEE Transactions on Vehicular Technology,2011,50(5):1267-1278.
[5]Tai J C,Tsang S T,Lin C P,et al.Real-time Image Tracking for Automatic Traffic Monitoring and Enforcement Applications[J].Image and Vision Computing,2004,22(6):485-501.
[6]马莹,吴楠楠.基于全球眼的地震现场应急救援交通路径分析系统研究[J].沈阳大学学报:自然科学版,2014,26(5):376-381.(Ma Ying,Wu Nannan.Earthquake Site Emergency Rescue Transportation Path Analysis System Based on Global Eye[J].Journal of Shenyang University:Natural Science,2014,26(5):376-381.)
[7]Pai C J,Tyan H R,Liang Y M,et al.Pedestrian Detection and Tracking at Crossroads[J].Pattern Recognition,2004,37(5):1025-1034.
[8]王自上.基于视频图像的交通事件自动识别算法研究[D].北京:北京交通大学,2010.(Wang Zishang.Automatic Recognition Algorithm Based on Video Images of Traffic Incidents[D].Beijing:Beijing Jiaotong University,2010.)
[9]刘晓男,李勃,陈启美,等.一种基于视觉技术的交通异常检测算法[J].仪器仪表学报,2011,32(S6):76-82.(Liu Xiaonan,Li Bo,Chen Qimei,et al.Vision Based Traffic Anomaly Detection Algorithm[J].Chinese Journal of Scientific Instrument,2011,32(S6):76-82.)
[10]邵士雨.基于视频的交通事件检测算法研究[D].济南:山东大学,2013.(Shao Shiyu.Research on Traffic Incident Detection Algorithm Based on Video[D].Jinan:Shandong University,2013.)
[11]Ki Y K,Lee D Y.A Traffic Accident Recording and Reporting Model at Intersections[J].IEEE Transactions on Intelligent Transportation System,2007,8(2):188-194.
[12]闻帆.基于视觉的交通路口车辆智能检测技术研究[D].哈尔滨:哈尔滨工业大学,2010.(Wen Fan.Research on Traffic Intelligent Detection Technology Based on Vision[D].Harbin:Harbin Institute of Technology,2010.)
[13]孙熙,李夏苗.基于boosting算法的交通事件检测[J].交通运输系统工程与信息,2007,7(5):37-41.(Sun Xi,Li Xiamiao.Traffic Incidents Detection Based on Boosting Method[J].Journal of Transportation Systems Engineering and Information Technology,2007,7(5):37-41.)
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心