基于计算机视觉和图像处理的交通参数检测

DETECTION OF TRAFFIC PARAMETERS BASED ON COMPUTER VISION AND IMAGE PROCESSING

摘要: 本文提出了一种基于计算机视觉和图像处理的车流量检测方法，通过分析CCDTV摄像机获取的图像中车辆和场景信息来有效地检测交通参数．本方法可以完成双向四车道高速公路和城市道路的车辆计数和车辆速度检测．实验结果表明：车辆计数的正确率为94%，车辆速度检测的正确率达92%，并具有满意的实时特性．

Abstract: This paper presents a method for vehicle speed detection based on computer vision and image processing.It can effectively detect vehicle speed by analyzing vehicle and scene information extracted from images.It merely needs two CCD TV cameras to complete road traffic parameter detection in a bi directional and four lane highway or an urban road.The experimental results show that the correct rates of vehicle count and vehicle speed detection are respectively 94% and 92%,and real time performance is satisfactory.