A Safety Evaluation of an Adaptive Traffic Signal Control System Using Computer Vision

The reliance on aggregate historical collision data as a sole technique in road safety analysis was proved challenging in the quest to better understand, predict, and improve road safety conditions. Therefore, surrogate safety measures such as the traffic conflict technique have been promoted as an alternative or complementary approach to assess and analyze road safety from a broader perspective than collision statistics alone. A primary focus of road safety analysis that could greatly benefit from vision-based road safety analysis is before-and-after (BA) evaluation of safety treatments. This study demonstrates the use of automated traffic conflict analysis in conducting a before-and-after (BA) safety study for an Adaptive Traffic Signal Control (ATSC) system. The objective of this study is to conduct a time-series (before-to-after) safety evaluation for two intersections in the City of Surrey where the ATSC system was implemented. The ATSC automatically makes real time adjustments to traffic signal timing based on actual observed traffic volumes to reduce vehicle delays and travel time. Overall, the study demonstrated the usefulness of using automated traffic conflicts in before-and-after safety evaluations of the ATSC system. Traffic conflicts occur more frequently than collisions so the desired sample size for analysis can be obtained in much shorter time periods. It was also demonstrated that the use of computer vision techniques to automate the extraction of traffic conflicts from video data can overcome the shortcomings of the traditional manual conflict observation methods. The results of the analysis showed considerable increase in the frequency and severity of conflicts following the implementation of the ATSC system. The increase of vehicle travel time following the implementation of the ATSC has likely contributed to the observed increase in conflict frequency and severity.