近年来交通拥堵成为目前各大城市面临的共同难题,本文通过机器学习算法进行车辆检测,根据车辆检测结果,计算道路的车辆密度和车辆速度,其中车辆速度是通过运动物体跟踪算法得到的,通过对历史交通数据的分析,根据车辆密度和车辆速度提出一种道路拥堵模型。车辆检测技术主要是使用Haar-like特征和Adaboost级联分类器。根据车辆检测结果和拥堵模型,判断当前路段的拥堵情况。 In recent years, traffic congestion has become a common problem faced by major cities. In this paper, vehicle detection is carried out by machine learning algorithm. Vehicle density and vehicle speed are calculated according to vehicle test results. The vehicle speed is obtained by moving object tracking algorithm. Based on the analysis of historical traffic data, a traffic congestion model based on vehicle density and vehicle speed is proposed. Vehicle detection technology is mainly using Haar-like features and Adaboost cascade classifier. According to the vehicle test results and congestion model, to determine the current section of the congestion situation.