An Adaptive Vehicle Detection Algorithm Based on Magnetic Sensors in Intelligent Transportation Systems
Bin Xu, Jianying Zheng, Qing Wang, Yang Xiao and Suat Ozdemir
Urban traffic congestion has become a very common phenomenon. In order to solve this problem, it is necessary to obtain the traffic data effectively. In this paper, we propose an adaptive vehicle detection algorithm based on magnetic sensors in intelligent transportation systems. The magnetic sensors used to detect the vehicles are deployed on the sides of roads to reduce the interference of sensors to the traffic other than the traditional methods in which sensors are often deployed on the traffic roads, and hence interfere with the normal traffic. In order to reduce the interference and improve the detecting accuracy rate, some kinds of effective algorithms to detect small signals are required. First of all, the deviation factor of the geomagnetic field signal is constructed to extract characteristics of the magnetic signals. Based on the definition, the y-axis of the magnetic signal is obtained and is proven to be the most obvious amplitude of variation. Based on this discovery, an adaptive vehicle detection algorithm is proposed. The proposed algorithm is tested on single lane and real and complicated urban road environments. Experimental results show that the proposed algorithm can achieve high detection accuracy.
Keywords: Intelligent transportation system, wireless sensor networks, vehicle detection, traffic congestion, magnetic sensor.