Research on Traffic Sign Area Detection and Classification Method Based on Improved YOLOv8s

DOI: https://doi.org/10.62517/jes.202602311

Author(s)

Feng Liu*, Weiwei Guo, Yunlong Zhao, Ying Li

Affiliation(s)

Department of Electrical and Information Engineering, Heilongjiang University of Technology, Jixi, Heilongjiang, China *Corresponding Author

Abstract

A detection and classification method based on improved YOLOv8s is proposed to address the problems of low recognition rate of traffic signs in complex environments, easy missed detection of small targets, and insufficient real-time performance. By integrating GTSRB, TT100K public datasets, and autonomously collected images, a dataset containing 45 categories and approximately 12000 samples was constructed through cleaning, enhancement, and unified annotation. Using YOLOv8s as the baseline, introduce CIoU loss function and adaptive anchor box clustering to optimize the confidence loss weight. The experiment shows that the improved method mAP@0.5 It reaches 73.9%, which is 7.1 and 8.0 percentage points higher than YOLOv8n and Faster R-CNN, respectively, with an inference speed of 120ms/frame. The confusion matrix and F1 curve validate the classification accuracy and anti-interference ability of the model, which can provide effective technical support for autonomous driving and intelligent traffic management.

Keywords

YOLOv8s; Object Detection; Deep Learning; Intelligent Transportation; Traffic Sign Detection

References

[1] Chen H H, Wang J Y, Ren X Y. A review of research on detection and recognition methods of traffic signs. Information Technology and Informatization, 2024, (3):77-82. [2] Zhou J W, Wang J P, Fu Y Y, etc. Traffic sign detection based on improved YOLOv5s. Journal of Anhui Engineering University, 2024, 39 (2): 40-46 [3] Wang H M, Zhou G D. Traffic Sign Recognition Based on YOLOv4 Model. Information Recording Materials, 2024, 25 (4): 61-63+66 [4] Wu Q. Design of Traffic Sign Detection Based on YOLO. Journal of Liaoning Transportation College, 2025, 27 (2): 27-31 [5] Ren S, He K, Girshick R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149. [6] Zheng K D, Dai D Z. Automated driving traffic sign detection algorithm based on improved YOLOv8. Information Technology and Informatization, 2025, (6):11-14. [7] Chen J X, Gan H Y. Traffic sign detection based on improved YOLOX. Practical Automotive Technology, 2024, 49 (2): 67-73 [8] Wang H, Zhang Q M, Gong D C. Small target detection algorithm for traffic signs in complex scenarios. Electronic Measurement Technology, 2025, 48 (2): 158-169 [9] Luo L, Xie Z K. Traffic sign detection algorithm based on improved YOLOv8. Mechanical and Electrical Engineering Technology, 2024, 53 (3): 205-210 [10] Cao L, Kang S B. Lightweight Traffic Sign Recognition Algorithm Based on FMS-YOLOv5s. Foreign Electronic Measurement Technology, 2024, 43 (5): 179-189.