Real-time detection and tracking of vehicle base fronts for measuring traffic counts and speeds on highways

A real-time system is presented for automatically monitoring a highway when the camera is relatively low to the ground and on the side of the road. In such a case, occlusion and the perspective effects due to the heights of the vehicles cannot be ignored. The system presented in this paper is based on a novel technique of detecting and tracking the base fronts of vehicles. Compared with the authors' previous approach, the proposed technique involves a simplified problem formulation, operates in real time, and requires only a small number of parameters to calibrate. Experimental results of a camera mounted at a height of 26 ft and a lateral distance of 12 ft from the edge of the pavement reveal accuracy as high as 98% with few false detections. The system can collect a variety of traffic data, including volume, time-mean speed and space-mean speed, density, vehicle classification, and lane change activities. By handling perspective effects and vehicle occlusions, the system overcomes some of the limitations of commercially available machine vision-based traffic monitoring systems that are used in many intelligent transportation systems applications.