Research on Recognition Technology of Vehicle Rolling Line Violation in Highway Based on Visual UAV
Diqing Zeng, Guigen Zeng, Prince Owusu Kodom · 2019
The recognition technology of vehicle illegal rolling line is an important part of intelligent transportation system. Aiming at the problems of traditional fixed monitoring and traditional rolling line detection algorithm, this paper designs a system model of vehicle illegal rolling line recognition based on visual UAV. Firstly, video data are collected by UAV patrol aerial photography, vehicle target detection is carried out by YOLOv3 algorithm based on deep learning, and then combined with lane line detection results based on closed angles average algorithm, according to whether the diagonal line of the bounding box of the vehicle intersects with the solid lane line to determine whether the vehicle is illegally compacted or not. The experimental results show that the accuracy and robustness of the proposed model are significantly improved compared with those of the traditional monitoring technology for illegal rolling. This provides an efficient and flexible monitoring mode for traffic monitoring and management.