Abstract:Traffic accident parameter determination is one of the most important topics in the traffic engineering domain. As a rule, parameter determination methods are all based on the brake marks, thrown objects etc. The results are often inaccurate due to the uncertainties of the road surface friction coefficient, the collision model and the geometric center of vehicles. In 2008, this study team proposed relying on large-scale construction of video surveillance equipment in China. A Gaussian mixture model is adopted to carry on background extraction and parameters such as vehicle movement path, real-time speed etc. were extracted from the video with Kalman Filter and obtained good results. Based on the research in 2008, this paper proposes that the background method of difference be used to replace the Gauss mixed model in the background extraction, and that a Kalman filtering track algorithm based on the fuzzy match be used to replace the traditional Kalman filtering track algorithm in the path track aspect. Through the theory, the example and the analysis, the newly proposed method is more effective.

Track Access Technology of Traffic Accident Vehicles Based on Video