This study set out to model the traffic noise level at intersections as a function of a set of variables from traffic characteristics, geometric features, pavement surface condition, and traffic management. A drone was used to take the video from traffic flow at the same time as measuring traffic noise level via a sound level meter. Regression modeling was used to determine the influential variables and their contribution to generated noise level at signalized T-intersections, cross-intersections, and roundabouts. The results showed that number of vehicles, and pavement condition index and speed together with the total traffic volume have significant impacts on the noise level. Furthermore, the traffic noise level at the exit approaches is higher than that at the entrance approaches of roundabouts. Moreover, the unexpected events and non-lane-based behavior caused a sudden increase in the maximum sound level at exit approaches.

这项研究着手对交叉路口的交通噪声水平进行建模，该水平是一组来自交通特征，几何特征，路面状况和交通管理的变量的函数。在通过声级计测量交通噪声水平的同时，使用无人机从交通流中获取视频。回归模型用于确定影响变量及其在信号T型交叉口，交叉口和回旋处产生的噪声水平的贡献。结果表明，车辆数量，路面状况指标和速度以及总交通量对噪声水平有显着影响。此外，出口处的交通噪声水平高于回旋处入口处的交通噪声水平。而且，