Traffic signs are an important component of road traffic systems.In practical applica-tions,traffic signs mainly exist outdoors and are easily affected by factors such as light,rain,and fog,resulting in uneven image acquisition quality.At the same time,due to factors such as shooting angle,distance,and focus,the obtained images are not high-definition enough,which poses a huge challenge to traffic sign detection.In order to solve the problem of traffic sign detection,based on deep reinforce-ment learning and combined with current classic object detection algorithms,in-depth analysis was conducted using YOLOv5 and YOL()v5-Tiny as basic network models.After algorithm improvement and optimization,the results showed that the accuracy of traffic sign detection was greatly improved,and it can accurately detect traffic signs in complex environmental conditions,and the detection has strong real-time performance and high practical value,which meets the requirements of algorithm ac-curacy for traffic sign detection.
关键词
深度强化学习/交通标识检测/算法
Key words
deep reinforcement learning/traffic sign detection/algorithm
维普
万方数据