Video-Based Analysis of Driving Safety of Electric Bicycles at Traffic Intersections

Abstract

This study presents a method for safety analysis to investigate the safety of electric bicycles at traffic intersections. Based on the development platform Visual Studio 2013, this proposed method uses the image processing functions from OpenCV library to track and detect the moving vehicles in videos of traffic intersections. Meanwhile, this paper puts forward a sieving algorithm based on the area factor to distinguish and recognize electric bicycles at traffic intersections. The traffic data of electric bicycles, such as speed and distance, can be obtained by processing video data. The distribution of electric bicycle's arrival speed and the rate of traffic interactions between electric bicycles and other vehicles are proposed to measure the severity of the intersections.