Vehicle tracking and traffic parameter extraction based on discrete wavelet transform

Real time traffic monitoring is one of the most challenging problems in machine vision. In this paper, we describe a method for multiple vehicles tracking and traffic parameters extraction based on Discrete Wavelet Transform. Occlusion is effectively resolved by our method. In order to describe the traffic characteristics, some important traffic parameters are extracted, such as vehicle trajectory, vehicle speed and vehicle count. Discrete wavelet transform is used to prevent the disturbance of fake motions. Experimental results are presented to demonstrate the effective tracking of vehicles on an urban artery.