TEJAS Journal of Technologies and Humanitarian Science

Satyam Gupta
Scholar, Department of Computer Science & Engineering (AI & ML) , KIPM College of Engineering and Technology, U.P., India

Author

Rupesh Prajapati
Scholar, Department of Computer Science & Engineering (AI & ML) , KIPM College of Engineering and Technology, U.P., India

Author

Ayush Raj
Scholar, Department of Computer Science & Engineering (AI & ML) , KIPM College of Engineering and Technology, U.P., India

Author

Shalini Yadav
Assistant Professor, Department of Computer Science & Engineering, KIPM College of Engineering and Technology, U.P., India

Author

📜 License:

This work is licensed under a Creative Commons Attribution 4.0 International License

• Share — Copy and Redistribute the material
• Adapt — Remix, Transform, and build upon the material
• The licensor cannot revoke these freedoms as long as you follow the license terms.

Abstract:

Urban intersections in India continue to rely on conventional fixed-timer traffic signals that operate without regard to the real-time density of vehicles on different lanes. These outdated systems contribute significantly to congestion, excessive fuel consumption, and inefficiencies in vehicle movement. To overcome these limitations, this research introduces an Intelligent Traffic Signal System (ITSS) that employs deep learning–based computer vision for automated and adaptive signal control. The framework integrates YOLO for high-speed vehicle detection, CNN-based vehicle classification, and LSTM-driven temporal traffic prediction to optimize signal duration based on density and flow patterns. Additional modules—including automatic challan generation, stolen vehicle identification using CNN, emergency vehicle prioritization, and lane obstruction detection—enhance traffic governance and enforcement capabilities. A diverse dataset compiled from Kaggle repositories, real-world city traffic recordings, and YouTube sources ensures.

References

[1] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You Only Look Once: Unified, Real-Time Object Detection,” Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2016, pp. 779–788.
[2] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” arXiv preprint arXiv:2004.10934, 2020.
[3] C.-Y. Wang, A. Bochkovskiy, and H.-Y. M. Liao, “YOLOv7: Trainable Bag-of-Freebies Sets New State-ofthe-Art for Real-Time Object Detectors,” arXiv:2207.02696, 2022.
[4] A. Krizhevsky, I. Sutskever, and G. Hinton, “ImageNet Classification with Deep Convolutional Neural Networks,” Communications of the ACM, vol. 60, no. 6, pp. 84–90, 2017.
[5] S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 17351780, 1997.
[6] J. Sun, Y. Xu, J. B. Wei, and L. Li, “Deep Learning-Based Traffic Flow Prediction Using LSTM Networks,” IEEE Transactions on Intelligent Transportation Systems, vol. 22, no. 7, pp. 4040–4051, 2021.
[7] H. Yaqoob, T. Abbas, and M. Khalid, “Vision-Based Vehicle Detection and Classification for Real-Time Traffic Monitoring,” IEEE Access, vol. 9, pp. 10262–10274, 2021.
[8] Y. Nishimura and M. Yamada, “Intelligent Traffic Signal Control Using LSTM Neural Networks,” 2020 IEEE 23rd International Conference on Intelligent Transportation Systems (ITSC), pp. 1–7.
[9] R. K. Satzoda and M. M. Trivedi, “Looking at Vehicles in the Night: Detection and Dynamics of Rear Lights,” IEEE Transactions on Intelligent Transportation Systems, vol. 13, no. 2, pp. 748–757, 2012.
[10] M. M. N. Rashid, M. Haque, and K. Hasan, “Automatic Traffic Violation Detection Using Deep Learning,” 2021 IEEE International Conference on Computer and Information Technology, pp. 521–526.