SURROGATE MEASURES OF SAFETY FOR SINGLE BICYCLE CRASHES
SMoS4SBC: Surrogate measures of safety for single bicycle crashes
Single bicycle crashes (SBCs), i.e., falls or impacts not involving contact with another road user, comprise over 80% of serious injuries to cyclists in Sweden, and historically these incidents have been a neglected as part of the road safety problem. Video-based analysis techniques have recently been developed to proactively identify infrastructure-related SBC risks through Surrogate Measures of Safety (SMoS).This project aims to extend and generalise this approach to allow for broader SBC risk assessments. Therefore, a cyclist tracking algorithm will be devised, combining computer vision techniques with mathematical modelling of bicycle/rider dynamics. The algorithm will be used infer safety-relevant cyclist movement details related to stability and dynamics from video data.
Key words: Single bicycle crashes, Surrogate measures of safety, Computer vision, Deep learning, Bicycle dynamics, Mathematical modelling
