Full Text PDF
Impact Factor (2025): 6.9
DOI Prefix: 10.47001/IRJIET
Impact Factor (2025): 6.9
DOI Prefix: 10.47001/IRJIET
Vol 10 No 5 (2026): Volume 10, Issue 5, May 2026 | Pages: 630-636
International Research Journal of Innovations in Engineering and Technology
OPEN ACCESS | Research Article | Published Date: 29-05-2026
Road traffic collisions are one of the main causes of avoidable deaths, and India alone saw 1.68 lakh road fatalities last year (2022). Manual surveillance techniques are inherently limited in scalability, and can be subject to human error. This paper proposes a complete automatic traffic violation detection and monitoring system specifically designed for Indian road conditions, which is cloud-based. This system combines YOLOv8 object detection, DeepSORT multi-object tracking and EasyOCR automatic number plate recognition to identify 5 important traffic violations: riding without helemt, triple seat riding, wrong side driving, red light jumping and overspeeding. In total, 15 video sequences (4.2 hours) were recorded under various environmental conditions, such as daylight, night, rain, and fog; the architecture was then tested using this set of videos. Ground truth annotations was verified using Cohen kappa coefficient, retaining only annotations with agreement above 0.80. Five-fold cross-validation yielded a macro-averaged F1-score of 90.7 percent at 29.0 frames per second on an NVIDIA T4 GPU. Automatic Number Plate Recognition accuracy reached 87.3 percent on Devanagari-Latin mixed plates following CLAHE-Gaussian preprocessing. Cloud load testing confirms linear scalability from one to fifty concurrent camera streams. The system has a Digital Personal Data Protection Act 2023 compliant governance framework covering data minimization, access control, and human-in-the-loop verification, providing a practical and privacy-aware foundation for smart-city traffic enforcement.
Detection of Traffic Violations, YOLOv8, DeepSORT, Automatic License Plate Recognition, EasyOCR, Computer Vision, AWS Cloud, Intelligent Transportation Systems, DPDPA 2023, Deep Learning.
Sandesh Shivaji Pawar, Priyanka Biswas, Sahil Bhima Durgude, & Gaurav Kumar Singh. (2026). Cloud-Native Intelligent Traffic Violation Detection and Monitoring System Using Deep Learning and Real-Time Video Stream Analysis. International Research Journal of Innovations in Engineering and Technology - IRJIET, 10(5), 630-636. Article DOI https://doi.org/10.47001/IRJIET/2026.105085
This work is licensed under Creative common Attribution Non Commercial 4.0 Internation Licence
Ministry of Road Transport and Highways, Govt. of India, "Road Accidents in India 2022," Transport Research Wing, New Delhi, 2023.
Government of India, "The Digital Personal Data Protection Act, 2023," Gazette of India Extraordinary, Part II, Section 1, New Delhi, 2023.
C. Stauffer and W. E. L. Grimson, "Adaptive Background Mixture Models for Real-Time Tracking," in Proc. IEEE CVPR, 1999, pp. 246-252.
J. Redmon et al., "You Only Look Once: Unified, Real-Time Object Detection," in Proc. IEEE CVPR, 2016, pp. 779-788.
Ultralytics, "YOLOv8: A New State-of-the-Art Model," 2023. Available: https://github.com/ultralytics/ultralytics
C.-Y. Wang and H.-Y. M. Liao, "YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information," arXiv:2402.13616, 2024.
A.Wang et al., "YOLOv10: Real-Time End-to-End Object Detection," arXiv:2405.14458, 2024.
Y. Zhao et al., "DETRs Beat YOLOs on Real-Time Object Detection," in Proc. IEEE CVPR, 2024, pp. 16965-16974.
A.Bewley et al., "Simple Online and Realtime Tracking," in Proc. IEEE ICIP, 2016, pp. 3464-3468.
N. Wojke et al., "Simple Online and Realtime Tracking with a Deep Association Metric," in Proc. IEEE ICASSP, 2017, pp. 3643-3647.
Y. Zhang et al., "ByteTrack: Multi-Object Tracking by Associating Every Detection Box," in Proc. ECCV, 2022, pp. 1-21.
N. Aharon et al., "BoT-SORT: Robust Associations Multi-Pedestrian Tracking," arXiv:2206.14651, 2022.
J. Cao et al., "Observation-Centric SORT," arXiv:2303.14260, 2023.
Y. Du et al., "StrongSORT: Make DeepSORT Great Again," IEEE Trans. Multimedia, vol. 25, pp. 8725-8737, 2023.
R. Smith, "An Overview of the Tesseract OCR Engine," in Proc. ICDAR, 2007, pp. 629-633.
J. Baek et al., "What is Wrong With Scene Text Recognition Model Comparisons?" in Proc. IEEE ICCV, 2019, pp. 4715-4723.
Y. Du et al., "PP-OCRv3: More Attempts for the Improvement of Ultra Lightweight OCR System," arXiv:2206.03001, 2022.
A.Sharma and R. Gupta, "Real-Time Traffic Violation Detection using AWS Kinesis Video Streams," Int. J. Adv. Comput. Sci. Appl., vol. 13, no. 4, 2022.
Ministry of Road Transport and Highways, Govt. of India, "e-Challan: Digital Traffic Enforcement," 2022. Available: https://echallan.parivahan.gov.in
L. Wen et al., "UA-DETRAC: A New Benchmark and Protocol for Multi-Object Detection and Tracking," Comput. Vis. Image Underst., vol. 193, 2020, 102907.
G. Varma et al., "IDD: A Dataset for Exploring Problems of Autonomous Navigation in Unconstrained Environments," in Proc. IEEE WACV, 2019, pp. 1743-1751.
A.Patel and S. Mishra, "Helmet Detection Dataset for Indian Traffic," Kaggle Dataset, 2022. Available: https://www.kaggle.com/datasets/