Robust vehicle detection and identification with single magnetic sensor

Abstract

Multiple vehicle detection and identification techniques have been widely applied to acquire road traffic information depending on various sensors, such as video camera, induction loop and magnetic sensor. Magnetic sensor can measure magnetic field distortion caused by the movement of vehicles and record reliable data. The precise volume and component of traffic flow can be obtained through reasonable data analysis. A novel vehicle detection algorithm is proposed based on the short-term variance sequence transformed from raw magnetic signal. The parking-sensitive module is introduced to enhance the robustness and adaptability of detection method. With abundant signal data, forty-two dimension features are extracted from every vehicle signal comprising statistical features of whole waveform and short-term features of fragment signal. The Gradient Tree Boosting algorithm is employed to identify four vehicle categories. The effectiveness of proposed approach is proved by the data collected at a freeway exit of Beijing. The vehicle detection algorithm shows an accuracy rate of 99.8 percent over 4507 vehicles and the vehicle identification algorithm shows an accuracy rate of 80.5 percent, which reveals enormous potential of field applications.