Research on Traffic Sign Recognition Algorithm Based on Deep Learning

View at publisher Summarise with VIRAT
Authors: Zhicheng He
Cited by: 0
References: 6

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper systematically studies the traffic sign recognition algorithm based on deep learning. Aiming at the problems of difficult small-object detection, strong environmental interference, and high real-time requirements in complex road environments, an improved YOLOv8-CA model is proposed. This model enhances the focusing ability on key features by introducing the Coordinate Attention mechanism, and combines the multi - scale feature fusion structure with the targeted data augmentation strategy, effectively improving the recognition performance and robustness of the model under complex conditions such as illumination changes, occlusions, and scale variations. Experiments on the TT100K dataset show that YOLOv8-CA improves the mAP by 2.6% compared with the baseline model, and the detection accuracy of small objects is increased by 4.2%. It exhibits better generalization ability under different background complexities and interference conditions, and at the same time maintains a high inference efficiency, making it suitable for real-time detection tasks in intelligent transportation systems.

Fields of Study & Subjects

Advanced Neural Network Applications Autonomous Vehicle Technology and Safety Computer Science and Engineering

Publication details

Journal / source
IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
Pages
754–759
Publication date
28 January 2026
Language
EN