RoadEye:- A Real Time Traffic Monitoring System Using AI

  This paper has presented Road Eye, a prototype AI-driven traffic monitoring and adaptive signal control system designed to address the fundamental inadequacies of fixed-time signal control in contemporary urban environments. The architecture integrates YOLOv8-based object detection, DeepSORT multi-object tracking, IoT sensor fusion, and a demand-responsive optimization module within a three-tier edge-to-server deployment model. The choice of edge computing as the primary inference platform is a deliberate architectural decision that resolves the latency constraints of cloud-dependent alternatives and simultaneously avoids the privacy risks associated with raw video transmission. These results position Road Eye as a technically viable and practically meaningful contribution to the field of Intelligent Transportation Systems.  

1. Asad, U., & Martin, Y. (2024). Latency optimization in edge-enabled urban traffic management: A comparative evaluation. Journal of Intelligent Transportation Systems, 18(4), 245–259.   2. Chen, X., Liu, Q., & Zhao, P. (2024). Comparative performance of single-stage object detectors in low-resource edge deployments for traffic surveillance. IEEE Transactions on Intelligent Transportation Systems, 25(3), 1124–1138.   3. Djahel, S., Salehie, M., & Lal, T. (2023). Environmental and economic costs of urban traffic congestion: A multi-city analysis. Transportation Research Part D: Transport and Environment, 119, 103–118.   4. INRIX. (2023). INRIX 2023 global traffic scorecard. INRIX Research.   5. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., & Berg, A. C. (2016). SSD: Single shot multibox detector. Computer Vision – ECCV 2016.   6. Liu, Z., Gao, Y., & Tian, K. (2023). Communication latency constraints in cloud-dependent intelligent traffic systems: Measurement and mitigation strategies. Vehicular Communications, 42, 100–117.   7. Lopez, P. A., et al. (2018). Microscopic traffic simulation using SUMO. In Proceedings of the 21st International Conference on Intelligent Transportation Systems.   8. Nguyen, T. H., & Park, J. W. (2024). Edge AI deployment strategies for real-time intersection monitoring: An empirical study. IEEE Access, 12, 34112–34128.   9. Redmon, J., & Farhadi, A. (2018). YOLOv3: An incremental improvement. arXiv preprint arXiv:1804.02767.   10. Schaffland, A., Nelson, J., & Schöning, J. (2024). Digital twin-enabled reinforcement learning for adaptive signal control in urban traffic networks. Transportation Simulation Journal, 9(1), 55–72.