Research on Traffic Incident Detection Algorithm Based On Video Image Analysis

摘要：

On the basis of video images from expressway monitoring camera and video vehicle recognition and tracking as well as the operation characteristic and management demand of the expressway in China, traffic flow parameters and traffic incident detection algorithms are researched on. Detection algorithm on speed, headway, occupancy and vehicle number, and automatic incident detection algorithm on parking, speeding, slow traffic, lane change, collision, traffic jam, traffic block and so on are proposed. Based on the above, the development of video detector of the AID system is implemented and applied in the expressway Management Monitoring System. Project application shows that the algorithms are feasible, direct and quick, and the recognition rate and reaction time meet the needs of expressway emergency management and rescue.

关键词： Video Image Video Recognition Automatic Incident Detection (AID)

作者: Xinsha Fu Juan Zhu

作者单位: School of Civil Engineering and Transportation South China University of Technology (SCUT) Guangzhou School of Computer Science & Engineering South China University of Technology (SCUT) Guangzhou, P.R.

会议类型: 国际会议

会议名称: 2010 International Conference on Measurement and Control Engineering(2010年IEEE测量与控制工程国际会议 ICMCE2010)

会议地点: 成都

会议语种:英文

页码: 700-704

在线出版日期: 2010-11-16（万方平台首次上网日期，不代表论文的发表时间）