Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM62687
Volume:
10 |
5 |
May 2026
Pages:
1-5
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Computer Science and Engineering - Artificial Intelligence and Machine Learning
Jain University
Computer Science and Engineering - Artificial Intelligence and Machine Learning
Jain University
Computer Science and Engineering - Artificial Intelligence and Machine Learning
Jain University
Computer Science and Engineering
Jain University
This paper presents a comprehensive, full-stack AI-powered platform for Smart Traffic Monitoring, Live Surveillance, and Automated Challan Generation. The system addresses critical limitations of conventional traffic enforcement — manual monitoring, delayed violation detection, fragmented data management, and slow reporting — by delivering an end-to-end automated pipeline suitable for smart-city deployment. The core AI pipeline integrates YOLOv8 for real-time vehicle detection and classification, DeepSORT for persistent multi-object tracking, a pixel-displacement speed estimation module, and EasyOCR combined with OpenCV preprocessing for license plate extraction. When a vehicle exceeds the configured speed threshold (80 km/h), the system automatically captures evidence images, reads the plate, and generates a digital challan with tiered fines (Rs. 1000 for 80–100 km/h; Rs. 2000 above 100 km/h). A FastAPI backend manages AI processing and relational persistence; a React + Vite frontend provides dashboards for upload management, live surveillance, violation review, and challan administration. Evaluation on real traffic videos demonstrates detection precision of 91.4%, tracking ID accuracy of 94.9%, speed estimation error of ±3.7 km/h, and OCR plate accuracy of 87.3%.
KEYWORDS
This paper presented the Smart Traffic Monitoring, Live Surveillance & Automated Challan System — a comprehensive, production-oriented traffic enforcement platform built on open-source technologies and deployable on standard hardware. The system delivers every component of the enforcement lifecycle, from raw video input to digital challan output, within a unified, modular, and maintainable codebase. The AI pipeline combining YOLOv8 detection, DeepSORT tracking, pixel-displacement speed estimation, and EasyOCR plate recognition provides reliable per-vehicle analysis under moderate traffic density and standard lighting, achieving 91.4% detection precision and 94.9% tracking ID accuracy. The modular architecture ensures that future improvements can be integrated without restructuring the core pipeline, establishing a strong and extensible foundation for smart-city traffic management.
1. Zhang, L., et al. (2022). Real-Time Traffic Monitoring and Anomaly Detection using YOLOv5 and Deep SORT. Journal of Applied Machine Vision. 2. Chen, Y., and Patel, S. (2023). A Comparative Analysis of Single-Stage Detectors for Urban Vehicle Classification. Proc. 2023 Int. Conf. on Computer Vision. 3. Kim, J., et al. (2023). Integrating OpenCV with Deep Learning for Directional Traffic Counting. Journal of Applied Computer Vision. 4. Gupta, V., and Ramu, K. (2020). Speed and Density Estimation using Perspective Transformation in Surveillance Videos. Journal of Intelligent Transportation Systems. 5. Wang, Z., and Li, S. (2023). Efficient Implementation of Multi-Object Tracking in Python using OpenCV. Int. Journal of Computer Vision and Pattern Recognition. 6. Ultralytics (2023). YOLOv8 Documentation. https://docs.ultralytics.com 7. FastAPI (2023). FastAPI Documentation. https://fastapi.tiangolo.com 8. JohannesStr, et al. (2021). Deep SORT: Simple Online and Realtime Tracking with a Deep Association Metric. GitHub Repository. 9. JaidedAI (2023). EasyOCR: Ready-to-use OCR with 80+ supported languages. https://github.com/JaidedAI/EasyOCR 10. React (2023). React Documentation. https://react.dev
Publication Date
5/6/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
8
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...