Traffic incident detection based on HMM

For an intelligent transportation system (ITS), traffic incident detection is one of the most important issues. In this paper, we propose a novel traffic incident detection method based on trajectory quantification and Hidden Markov Model (HMM) classifier. First, object detection algorithm that combines geodesic active contour model based on level set theory and background subtraction was proposed and accurate contour of moving object is got. Sencondly, the kalman filter is applied to predict the possible trajectories of moving object and then trajectory feature was extracted as HMM input. Finally, HMM was used for classification of U-turns, illegal turn left, illegal change lanes. The experimental result showed that the method proposed has better robustness and higher recognition rate.