夜间车流量检测算法

Night traffic detection algorithm

为了解决夜间车辆检测和车辆跟踪的问题,提出了一种基于车前灯的夜间车流量检测算法。首先,分析了夜间车辆检测的特点及采用车前灯进行夜间车辆检测的原因。其次,利用图像二值化的方法对车灯进行分割,通过比较车灯和车灯投射光线之间的差异识别出车灯,然后在此基础上分别利用车灯帧内配对规则和车灯帧间匹配规则实现车辆检测和车辆跟踪。最后,在跟踪过程中,利用该算法实现了车流量的统计。

In order to solve the problem of nighttime vehicle detection and vehicle tracking,proposes a nighttime traffic detection algorithm based on headlights.Firstly,analyzed the features of nighttime vehicle detection and using the headlights of the vehicle to detect the reasons for night.Secondly,Using image binaryzation method to segmentation for headlights,by compare the difference between light and the light projection light recognition lights,then on this basis,using light frame matching rules and matching rules between light frame respectively to realize vehicle detection and vehicle tracking.Finally,in the process of tracking,using the algorithm to realize the traffic statistics.