Vision-based real-time traffic accident detection

Recently, traffic accident detection is becoming one of the interesting fields due to its tremendous application potential in Intelligent Transportation Systems. In this paper, we present a vision-based real time traffic accident detection method. We intend to extract foreground and background from video shots using the Gaussian Mixture Model (GMM) to detect vehicles; afterwards, the detected vehicles are tracked based on the mean shift algorithm. Then the three traffic accident parameters including the changes of the vehicles position, acceleration, and the direction of the moving vehicles are gathered to make the final accident decision. The experimental results on real video demonstrate the efficiency and the applicability of the proposed approach.