Algorithm research for automatic incident detection based on wavelet analysis and contour method

Abstract

Automatic incident detection algorithm is the core of freeway traffic incident management. In this paper, the wavelet analysis and contour method are used to present the automatic traffic incident detection algorithm, which detects the singularity of traffic parameters signal and estimates the time of the traffic events. The algorithm is validated by the simulation data of Microscopic traffic simulation system VISSIM. The result shows that the algorithm has good applicability, accuracy and reliability, with good practical value.