Research on traffic flow detection based on YOLOv5
The principle of target detection and traffic detection based on convolutional neural network is studied,and YO-LOv5,which has good performance in real-time and accuracy,is selected as the target detection model.Then label the collected traffic video data with Labelme,and program with PyCharm Community Edition 2020 as the experimental environment to realize the two main functions of vehicle target detection and flow detection.The test results show that the video traffic flow detection soft-ware based on YOLOv5 can accurately detect the current traffic flow no matter in the daytime or at night under various interference conditions,and has strong robustness,real-time and accuracy,which can meet the needs of practical application.
NETL
NSTL
万方数据
