Speed Analysis Based on Engine Sound of Driving Vehicle

Abstract

Vehicle speed at the time of an accident has a significant impact on the reliability of the reconstruction of the traffic accident. Recently, vehicle speed analysis methods, based on acoustic data from cameras, have been studied to overcome the limitation of image data due to blind spots and environmental conditions. In this paper, we proposed a speed analysis technique based on external noise and the Doppler effect generated by a moving vehicle. In particular, the frequency of vehicle engine sound was analyzed by using STFT, and the speed of the vehicle with respect to time was estimated by correcting the two-dimensional Doppler effect. Finally, the reliability of the proposed method was verified through acoustic simulation by using Matlab and actual vehicle evaluation by using acoustic data from dashboard cameras. Therefore, the proposed technique that utilizes acoustic data from cameras under limited conditions could estimate the speed of the vehicle in an accident.