Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (34): 239-242.DOI: 10.3778/j.issn.1002-8331.2008.34.072

• 工程与应用 • Previous Articles     Next Articles

Vehicle breaking detection on express way based on particle filter algorithm

SHI Shi-xu1,2,ZHENG Qi-lun1,HUANG Han1,JIA Xi-ping1   

  1. 1.College of Computer Science and Engineering,South China University of Technology,Guangzhou 510640,China
    2.Guangdong Xinyue Communication Investment Corporation Limited,Guangzhou 510410,China
  • Received:2007-12-10 Revised:2008-02-18 Online:2008-12-01 Published:2008-12-01
  • Contact: SHI Shi-xu

基于粒子滤波算法的高速公路车辆停车检测

石时需1,2,郑启伦1,黄 翰1,贾西平1   

  1. 1.华南理工大学 计算机科学与工程学院,广州 510640
    2.广东新粤交通投资有限公司,广州 510410
  • 通讯作者: 石时需

Abstract: A method of vehicle breaking detection based on particle filter is proposed,for automatic detection of traffic video on express way.A hybrid difference approach is used to obtain the vehicle objects of video.An improved particle filter algorithm is designed for tracking the vehicle of the video.The mathematic model of moving and breaking behavior of the tracking vehicle is given for detecting.Finally,several traffic videos are tested,and the results indicate the proposed method is efficient,high-detection-rate and robust.

Key words: particle filter, video tracking, vehicle breaking detection, video incident detection

摘要: 提出了一种基于粒子滤波视频跟踪算法的停车事件检测方法,实现了对高速公路交通视频的自动监控。首先用混合差分技术,快速提取出视频中的车辆对象;并用粒子滤波算法实现了运动车辆的跟踪;进而通过对车辆运动的数学建模,对停车事件进行了自动检测。最后,对多组高速公路交通视频进行测试,结果表明:提出的检测方法比其他常用方法响应速度更快,且具有较高的检测准确率和鲁棒性。

关键词: 粒子滤波, 视频跟踪, 停车检测, 视频事件检测