Traffic sign recognition plays a vital role in intelligent transportation systems, enabling driver assistance systems to effectively reduce traffic accidents. In this paper, a traffic sign recognition system based on an improved version of YOLOv5s was developed. Firstly, the addition of a channel attention module (CBAM) enhanced the network’s ability to extract informative features from the input data in both spatial and channel dimensions. Then, the original target box in YOLOv5s network which was not suitable for this detection task is improved, and the clustering method of target box obtained from the original network was optimized to K-Means++ clustering method. The researchers trained this method using the CSUST Chinese Traffic Sign Detection Benchmark (CCTSDB) Dataset and achieved promising results in recognizing traffic signs. The experimental results demonstrated that the improved YOLOv5s outperformed YOLOv5s, achieving a Precision of 97.8% and an mAP of 98.8%.