Authors : Prachi Sorte; Suchit Eknath Bodake; Pratiksha Gorakshnath Todmal

Volume/Issue : Volume 10 - 2025, Issue 3 - March

---

Google Scholar : https://tinyurl.com/y4c2ra44

Scribd : https://tinyurl.com/bdd529zb

DOI : https://doi.org/10.38124/ijisrt/25mar1744

Google Scholar

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 15 to 20 days to display the article.

---

Abstract : Traffic sign detection and recognition are crucial for intelligent transportation systems (ITS) and autonomous vehicles. Traditional methods face challenges with distorted or unclear images, impacting detection accuracy. This research proposes a real-time traffic sign detection system using YOLOv4 integrated with Generative AI (GenAI) to correct distorted images and enhance contextual data readability. By leveraging GenAI, the model achieves improved detection accuracy and robustness, particularly under adverse conditions. Experimental results demonstrate an accuracy of 89.33%, showcasing the system’s efficiency and real-time performance.

Keywords : Traffic Sign Detection, YOLOv4, Generative AI, Real-Time Recognition, Contextual Data Enhancement.

References :

1. C. Dewi, R.-C. Chen, Y.-T. Liu, X. Jiang, and K. D. Hartomo, “YOLO V4 for Advanced Traffic Sign Recognition With Synthetic Training Data Generated by Various GAN,” IEEE Access, vol. 9, pp. 97228–97240, Jul. 2021. doi: 10.1109/ACCESS.2021.3094201​
2. J. Yang, T. Sun, W. Zhu, and Z. Li, “A Lightweight Traffic Sign Recognition Model Based on Improved YOLOv5,” IEEE Access, vol. 11, pp. 115998–116010, Oct. 2023. doi: 10.1109/ACCESS.2023.3326000​
3. S. Khalid, J. H. Shah, M. Sharif, F. Dahan, R. Saleem, and A. Masood, “A Robust Intelligent System for Text-Based Traffic Signs Detection and Recognition in Challenging Weather Conditions,” IEEE Access, vol. 12, pp. 78261–78274, Jun. 2024. doi: 10.1109/ACCESS.2024.3401044​
4. S. Luo, C. Wu, and L. Li, “Detection and Recognition of Obscured Traffic Signs During Vehicle Movement,” IEEE Access, vol. 11, pp. 122516–122528, Nov. 2023. doi: 10.1109/ACCESS.2023.3329068​
5. Y. Wang and Q. Wang, “Robust Stacking Ensemble Model for Traffic Sign Detection and Recognition,” IEEE Access, vol. 12, pp. 178941–178953, Dec. 2024. doi: 10.1109/ACCESS.2024.3504827​
6. X. Yuan, A. Kuerban, Y. Chen, and W. Lin, “Faster Light Detection Algorithm of Traffic Signs Based on YOLOv5s-A2,” IEEE Access, vol. 11, pp. 19395–19407, Mar. 2023. doi: 10.1109/ACCESS.2022.3204818​
7. M. Lopez-Montiel, U. Orozco-Rosas, M. Sánchez-Adame, K. Picos, and O. H. Montiel Ross, “Evaluation Method of Deep Learning-Based Embedded Systems for Traffic Sign Detection,” IEEE Access, vol. 9, pp. 101217–101229, Jul. 2021. doi: 10.1109/ACCESS.2021.3097969​
8. Q. Gao, H. Hu, and W. Liu, “Traffic Sign Detection Under Adverse Environmental Conditions Based on CNN,” IEEE Access, vol. 12, pp. 117572–117584, Sep. 2024. doi: 10.1109/ACCESS.2024.3446990​
9. W. H. D. Fernando and S. Sotheeswaran, “Traffic Sign Recognition Using Convolutional Neural Network and Skipped Layer Architecture,” IEEE Conference Proceedings, 2021. doi: 10.1109/Conference.2021​
10. J. Yang, T. Sun, W. Zhu, and Z. Li, “A Lightweight Traffic Sign Recognition Model Based on Improved YOLOv5,” IEEE Access, vol. 11, pp. 115998–116010, Oct. 2023. doi: 10.1109/ACCESS.2023.3326000​
11. Z. Charouh, A. Ezzouhri, M. Ghogho, and Z. Guennoun, "Video Analysis and Rule-Based Reasoning for Driving Maneuver Classification at Intersections," IEEE Access, vol. 10, pp. 45102–45112, May 2022. doi: [10.1109/ACCESS.2022.3169140]​
12. R. Suresha, N. Manohar, G. A. Kumar, and M. R. Singh, "Recent Advancement in Small Traffic Sign Detection: Approaches and Dataset," IEEE Access, vol. 12, pp. 192840–192847, Dec. 2024. doi: [10.1109/ACCESS.2024.3514692].
13. T. Wang, J. Zhang, B. Ren, and B. Liu, "MMW-YOLOv5: A Multi-Scale Enhanced Traffic Sign Detection Algorithm," IEEE Access, vol. 12, pp. 148880–148884, Oct. 2024. doi: [10.1109/ACCESS.2024.3476371]​