for Automatic Traffic Surveillance

Abstract

In this paper, a system for wavelet-based vehicle tracking for automatic traflFic surveillance is proposed. In order to meet real-time requirements, we use an adaptive threshold@ and Wavelet-based Neural Network (NN), which achieves low computational complexity, accuracy of localization, and noise robustness has been considered for vehicle tracking. The proposed system consists of three steps: moving region extraction, vehicle recognition and vehicle tracking. First, moving regions are extracted by performing a frame difference analysis on two consecutive frames using adaptive thresholding. Second, the Wavelet-based NN is used for recognizing the vehicles in the extracted moving regions. Wavelet Transform (WT) is adopted to decompose an image and a particular frequency band is selected for input of NN for vehicle recognition. Third, vehicles are tracked by using position co-ordinates and wavelet features difference values for correspondence in recognized vehicle regions. Experimental results of the proposed system can be useful for applying traffic surveillance system.