Traffic sign detection algorithm based on YOLOv5 combined with BIFPN and attention mechanism
Xiangyu Zhang, Shaoqing Mo, Zhaobao Wan · 2023
With the development of autonomous driving technology and advanced assisted driving, the task of traffic sign detection is becoming more and more important. However, due to the complex scene of traffic signs and small targets, the detection accuracy is low and the false detection rate is high. For this reason, an improved YOLOv5 algorithm model is proposed. The algorithm introduces the CBAM attention mechanism module to improve the feature extraction ability of the network model in complex scenes, and integrates the bidirectional Feature Pyramid Network to achieve multi-scale feature fusion. The method is compared with the original YOLOV5, Faster-RCNN, and other classic algorithms in the field of target detection on the same data set. The detection effect is better in the scene of night vision and occlusion of the target. The mAP value of the improved YOLOv5 has increased by 2.1% to 91.6%.