Vehicle Detection Device Using 2-Axis Magneto-Impedance Sensors for Traffic Monitoring

This study proposes a method for measuring moving vehicles using dual-axis high-precision magneto-impedance (MI) sensors. The study investigates the impact of various experimental factors on the measurement results and establishes an optimal experimental detection method. Two sets of dual-axis MI sensors were utilized to form a car measurement device, and a data processing algorithm that enables real-time vehicle detection, traffic congestion detection, vehicle speed estimation, and vehicle type identification is presented. In random measurements of 649 cars, a detection accuracy of 91.8% is achieved. The speed measurement accuracy for individual vehicles reaches 98.3%. The challenges in using magnetic sensors for vehicle detection and their applications in deep learning are also discussed in this study.