A Robust Fusion Method for Vehicle Detection in Road Traffic Surveillance

Vehicle detection is an essential task in the intelligent transportation system, which will affects the performance of surveillance directly. This paper presents an approach to detect vehicle from a sequence of traffic images obtained from expressway scenes. Firstly, inter-frame difference method was used to choose some frames with small traffic flow, and then pixels detected as background are being used to initialize background by calculating average value. Secondly, the vehicle is detected by fusing inter-frame difference, background subtraction and edge-based background subtraction methods together. Finally, the vehicle region can be obtained by implementing morphological processing. Meanwhile, the pixels detected as background were being used to update the background. The experimental results from highway scenes show that the algorithm is effective.