YOLOV5-based traffic sign detection algorithm

For the problems of low recognition accuracy and slow detection speed in traffic sign recognition tasks, an improved YOLOv5-based traffic sign recognition model is proposed. The first step is to replace Shufflenetv2 with a lightweight network. Firstly, Shufflenetv2, a lightweight network, is used to replace the YOLOv5 backbone network to improve the detection speed of the model; then, BiFPN is used as the feature fusion structure in the Neck layer to achieve multi-scale fusion; finally, K-means algorithm is used to reacquire the initial anchor frame value of the model. The experimental results show that the recognition accuracy of the improved network model is better than that of the original YOLOv5, and the recognition of traffic signs is improved.