摘要
为了方便快速地进行车速测量,提出了一种用于测量视频图像中车辆速度的方法。该方法取运动角点为特征量,通过选择灰度相关函数为特征匹配函数,实现了车辆在一定帧差内移动距离的测量。与传统方法相比,该方法不仅对设备的限制更小,而且运算速度更快。
This paper presents a method to measure speed of vehicles in video sequences conveniently and rapidly. Motion corner points are chosen to form a feature vector, and the image intensity is used as a feature matching function to measure inter-frame movement of the vehicle. With the present method, device limitations are alleviated and processing time reduced compared to traditional techniques.
出处
《中国图象图形学报》
CSCD
北大核心
2006年第11期1650-1652,共3页
Journal of Image and Graphics
关键词
HARRIS角点检测
运动角点
灰度相关
Harris corner detector, motion corner point, intensity correlation