Traffic sign recognition serves a core technology in intelligent transportation systems and autonomous driving, with its performance directly impacting road safety and traffic efficiency. Addressing the limitations of existing methods in complex scenario adaptation and small object detection accuracy, this paper proposed an enhanced traffic sign detection model based on modified YOLOv8 architecture. Performance improvements were realized through three architectural innovations: Shift-ConvNets modules was incorporated to improve local feature extraction, the Auxiliary Head structure is developed to refines multi-scale feature fusion, the Focal Modulation was integrated to elevate small target detection sensitivity. Experimental evaluation is conducted on the standard CCTSDB2021 dataset. The improved model achieves 80.41% mAP50 and 73.69% Recall, representing significant improvements of 2.54% and 2.27% over the baseline YOLOv8n model. The effectiveness of each improved module and their synergistic optimization mechanism were verified through systematic ablation experiments. This Research provides a practical solution to current technical challenges in traffic sign recognition. The proposed methodology offer valuable insights for optimizing environmental perception modules in intelligent transportation systems and autonomous driving.
Research on traffic sign recognition based on YOLOv8
Fifth International Conference on Optical Imaging and Image Processing (ICOIP 2025) ; 2025 ; Xi’an, China
Proc. SPIE ; 13688 ; 1368824
2025-07-28
Conference paper
Electronic Resource
English
Traffic sign recognition system based on YOLOv8-ConvNeXt
SPIE | 2025
|Traffic Sign Detection Using YOLOv8
IEEE | 2024
|YOLO-TSR: A Novel YOLOv8-Based Network for Robust Traffic Sign Recognition
Transportation Research Record | 2025
|