摘要
车辆视频检测系统作为智能交通系统的重要组成部分,为交通控制与交通规划提供必要的数据来源。考虑到外界环境的影响以及数据传输过程中的干扰,对所拍摄的单一图像的分析难以获得准确信息。因此,在简要介绍D-S证据理论的基础上,在对由摄像机拍摄的交通现场画面做了图像处理后,提取出车辆特征,并最终运用D-S证据理论来进行数据融合,从而来提高交通视频检测系统中车辆类型的识别率。实验证明采用此方法将使得车辆识别率大为提高。
As one of the important parts of the intelligent transport system, vehicle video detection system can provide information for traffic control and traffic programming. It' s difficult to obtain accurate information from single video image, hecause of the influence of weather condition and data transmission. This paper presents briefly the con- cepts of D- S evidence theory, extracts vehicle characters from the processed images obtained by camcorder, and then explains how to improve the accuracy of traffic parameters by means of D - S evidence theory. Experimental result shows that the method can improve the accuracy of vehicle recognition.
出处
《计算机仿真》
CSCD
2007年第7期264-267,共4页
Computer Simulation
关键词
证据理论
视频检测
图像处理
车辆识别
Evidence theory
Video detection
Image processing
Vehicle recognition