A novel method of predictive collision risk area estimation for proactive pedestrian accident prevention system in urban surveillance infrastructure

Abstract

Road traffic accidents, especially vehicle-pedestrian collisions in crosswalk, globally pose a severe threat to human lives and have become a leading cause of premature deaths. In order to protect such vulnerable road users from collisions, it is necessary to handle possible conflict in advance and warn road users, not post-facto. A breakthrough for proactively preventing pedestrian collisions is to recognize pedestrian's potential risks based on vision sensors such as CCTVs. In this study, we propose a predictive collision risk area estimation system at unsignalized crosswalks. The proposed system applied trajectories of vehicles and pedestrians from video footage after preprocessing, and then predicted their trajectories by using deep LSTM networks. With use of predicted trajectories, this system can infer collision risk areas statistically, further severity of levels is divided as danger, warning, and caution. In order to validate the feasibility and applicability of the proposed system, we applied it and assessed the severity of potential risks in two unsignalized spots with different mobility environment in Osan City, Republic of Korea. As a result, the ratio of dangerous scenes is higher in Spot A (0.115) than in Spot B (0.077) when applying the best performance model in each spot, and we found that the risk situation varies depending on the mobility environment.