Development of Image Sequences Based Traffic Incident Detection System for Urban Intersection

Traffic incident detection is one of the most important issues for intelligent transportation systems (ITS), especially in urban area which is full of signaled intersections. This paper presents the development of a novel traffic incident detection system based on image signal processing, feature extraction algorithms, and hidden Markov model (HMM) classifier. First, a traffic surveillance system was set up at a typical intersection of china, traffic videos were recorded and image sequences were extracted for image database forming. Second, several features extraction algorithms were used and compared. Finally, HMM was used for classification of traffic signal logics (East-West, West-East, South-North, North-South) and accident of crash. Feature generation with DCT-FFT process gives the best result with total correct rate of 91% and incident recognition rate of 95%.