This paper proposes an enhanced YOLO network for improving the efficiency of traffic sign detection. The research focuses on a comprehensive set of 72 distinct traffic signs prevalent on urban roads in China. The enhanced model modifies the backbone network by removing the terminal convolution module and Conv3 (C3) module, replacing 32-fold downsampling with 16-fold downsampling, and introducing a 152×152 feature fusion module. A novel hybrid spatial pyramid pooling module, H-SPPF, is introduced to capture more contextual information. A channel attention mechanism is also integrated with three other improved methodologies. The enhanced algorithm achieves a precision rate of 91.72%, a recall rate of 91.77%, and a mean average precision (mAP) of 93.88% at an intersection over union (IoU) threshold of 0.5, and an mAP of 75.81% for a range of IoU criteria between 0.5 and 0.95. The method is validated on an augmented dataset established for this study. The paper also presents an improved K-means++ anchor frame clustering algorithm, a refined backbone network structure and H-SPPF module, and a channel attention mechanism. The enhanced model demonstrates improved performance in detecting small traffic signs, with significant improvements in accuracy and efficiency. The results show that the proposed model outperforms the standard YOLOv5s model in various scenarios, including challenging conditions such as dim lighting and interference from surrounding advertisements. The model's ability to accurately detect tilted traffic signs and smaller distant signs is especially impressive, demonstrating its advanced characteristics that outperform those of the YOLOv5s model. The study concludes that the proposed enhancements significantly improve the accuracy of detecting small traffic signs and demonstrate exceptional resilience in a range of difficult situations. This research builds a solid foundation for future developments in practical applications, like autonomous driving and intelligent transportation systems.This paper proposes an enhanced YOLO network for improving the efficiency of traffic sign detection. The research focuses on a comprehensive set of 72 distinct traffic signs prevalent on urban roads in China. The enhanced model modifies the backbone network by removing the terminal convolution module and Conv3 (C3) module, replacing 32-fold downsampling with 16-fold downsampling, and introducing a 152×152 feature fusion module. A novel hybrid spatial pyramid pooling module, H-SPPF, is introduced to capture more contextual information. A channel attention mechanism is also integrated with three other improved methodologies. The enhanced algorithm achieves a precision rate of 91.72%, a recall rate of 91.77%, and a mean average precision (mAP) of 93.88% at an intersection over union (IoU) threshold of 0.5, and an mAP of 75.81% for a range of IoU criteria between 0.5 and 0.95. The method is validated on an augmented dataset established for this study. The paper also presents an improved K-means++ anchor frame clustering algorithm, a refined backbone network structure and H-SPPF module, and a channel attention mechanism. The enhanced model demonstrates improved performance in detecting small traffic signs, with significant improvements in accuracy and efficiency. The results show that the proposed model outperforms the standard YOLOv5s model in various scenarios, including challenging conditions such as dim lighting and interference from surrounding advertisements. The model's ability to accurately detect tilted traffic signs and smaller distant signs is especially impressive, demonstrating its advanced characteristics that outperform those of the YOLOv5s model. The study concludes that the proposed enhancements significantly improve the accuracy of detecting small traffic signs and demonstrate exceptional resilience in a range of difficult situations. This research builds a solid foundation for future developments in practical applications, like autonomous driving and intelligent transportation systems.