Traffic Flow Measuring System by Image Processing

In this paper, we show anew image sensor for measuring traffic flow. The image sensor is set at an intersection. The image sensor measures t r af f i c flow by vehicle tracking. The tracking is done by gray level pattern matching. Gray level pattern matching is suitable for outdoor use since it is robust against change of scene brightness. Then we provide the hardware for high speed gray level pattern matching. After showing our new method for measuring traffic flow by the image sensor, we measure traff~c flow at an actual intersection and describe the results. The sensor measures traffic flow accurately under various conditions. INTRODUCTION To resolve t mf 3c congestion and to make good use of roads, several traffic signal control systems[l] and traftic information service systems[2] have been developed. In traffic control systems, traffic signals at several intersections are controlled cooperatively so as not to localize traffic in certain areas. Traffic information service systems inform divers of td5c situations to prevent traffic congestion from spreading. For these systems dynamic traffic information is necessary. An image sensor is one of the most important techniqes to get the information at many road points.