Development of new methods for estimating/ improving pedestrian safety based on video analytics

该项目的目的是继续开发数学关系，将行人和车辆的碰撞风险与交通特征和建筑环境特征联系起来。建立车辆与车辆、车辆与行人冲突或未遂事故以及此类碰撞之间的量化关系将允许仅需要快速获取冲突数据而不需要多年的碰撞数据进行碰撞预测。碰撞记录需要多年的积累，因为碰撞是罕见的事件，并且建筑环境在这个收集期间经常发生变化。这些事件的自我报告也存在较高程度的主观性。该项目旨在建立的关系将使风险预测变得更容易、更准确，从而产生经济和社会效益。

The purpose of this project is to continue the development of mathematical relationships that link collision risk for pedestrians and vehicles to traffic characteristics and features of the built environment. Establishing a quantified relationship between vehicle-vehicle and vehicle-pedestrian conflicts, or near misses, and collisions of this type will allow for collision forecasts that only require quickly obtained conflict data as opposed to requiring years of collision data. Collision records require years of accumulation, since collisions are rare events, and the built environment often changes within this collection period. There is also a higher degree of subjectivity in the self-reporting of these events. The relationships this project aims to build will allow for easier and more accurate risk forecasting resulting in economic and societal benefits.