Smart Traffic Management System Based on Object Detection by Using YOLO

The development of the proposed Smart Traffic Management System demonstrates the practical application of computer vision and intelligent decision-making techniques in solving urban traffic challenges. By integrating real-time vehicle detection and adaptive signal control, the system provides an automated approach for managing varying traffic conditions more efficiently than conventional methods. The implementation validates the feasibility of combining detection and prediction mechanisms within a unified framework.   Experimental observations indicate that the system can successfully detect vehicles, estimate traffic density, and dynamically allocate signal durations according to congestion levels. The adaptive behavior of the system improves traffic movement by prioritizing highly congested lanes while maintaining balanced signal distribution. The modular design also enables stable operation and supports future expansion without major architectural changes.   The proposed framework establishes a foundation for next-generation intelligent transportation systems and contributes toward smart city development. Future improvements can include integration of emergency vehicle prioritization, multi-intersection coordination, IoT connectivity, and predictive traffic analytics to create a more advanced and fully autonomous traffic management ecosystem.

[1]  S. R. Joshi, U. Rakesh and S. Sripad, “Exploring Sustainable Techniques for Traffic Management in Urban Environments,” 2024 International Conference on Recent Innovation in Smart and Sustainable Technology (ICRISST), IEEE, 2024.   [2]   B. Liu, C.-T. Lam, B. K. Ng, X. Yuan and S. K. Im, “A Graph-Based Framework for Traffic Forecasting and Congestion Detection Using Online Images from Multiple Cameras,” IEEE Access, vol. 12, pp. 3756–3772, Jan. 2024, doi: 10.1109/ACCESS.2023.3349034.   [3]    C. Güven, Ö. Karaduman and E. Avcı, “A Review on Urban Intelligent Traffic Management Problems: Sensors and Methods,” 2021 2nd International Informatics and Software Engineering Conference (IISEC), IEEE, 2021, doi: 10.1109/IISEC54230.2021.9672412.   [4]   R. Song and H. Yang, “Clustering and Understanding Traffic Flow Patterns,” IEEE Conference Publication, IEEE, 2021.   [5]    A. Chowdhury et al., “Priority Based and Secured Traffic Management System for Emergency Vehicles using IoT,” International Conference on Engineering, Management, Industrial and Scientific Innovation Systems, IEEE, 2016.   [6]   C. Liu and H. Yang, “Toward a Dynamic Reversible Lane Management Strategy by Empowering Learning-Based Predictive Assignment Scheme,” 2022 IEEE Smart World Congress (SWC), 2022, doi: 10.1109/SWC57546.2023.10448974.   [7]   Z. Wei, T. Peng and S. Wei, “A Robust Adaptive Traffic Signal Control Algorithm Using Q-Learning under Mixed Traffic Flow,” Sustainability, vol. 14, no. 10, Article 5751, 2022, doi: 10.3390/su14105751.   [8]   “Adaptive Urban Traffic Signal Control for Multiple Intersections: An LQR Approach,” 2022 25th IEEE International Conference on Intelligent Transportation Systems (ITSC), Macau, China, 2022, doi: 10.1109/ITSC55140.2022.9922033