Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{198677,
        author = {KAILASAPRABHU.C and LOKESH A and RANJITHKUMAR S and VASANTH S and PONNEELA VIGNESH R},
        title = {YOLO-Based Smart Traffic Violation Detection and Accident Monitoring System},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {9842-9848},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=198677},
        abstract = {Traffic rule violations and road accidents constitute a serious global challenge, resulting in millions of deaths and substantial economic losses annually. Traditional traffic monitoring systems reliant on manual observation and static CCTV cameras suffer from human error, limited coverage, and a lack of proactive enforcement, often acting reactively rather than preventively. This project introduces an automated, real-time system for detecting traffic violations and accidents, utilizing the state-of-the-art YOLOv12 object detection architecture. The proposed system is capable of detecting various traffic violations, including the failure of two-wheeler riders to wear helmets, triple riding, and license plate identification, while also identifying accident incidents through vehicle trajectory analysis and collision detection algorithms. The YOLOv12 model achieves superior detection accuracy and faster inference speeds compared to previous YOLO versions, making it highly suitable for real-time deployment. A comprehensive dataset of traffic scenarios was curated and annotated specifically for training purposes. The system integrates a Flask-based web application that provides user authentication, real-time video stream processing, batch image analysis, and detailed violation reporting, complete with timestamps and geolocation data. Detection history is stored in an SQLite database for subsequent auditing and analysis. On a standard hardware configuration, the proposed system achieved a Mean Average Precision (mAP) of 98.7% on a test dataset, demonstrating a real-time processing capability of over 45 frames per second (FPS); this offers a scalable solution for Intelligent Traffic Management Systems within smart city environments.},
        keywords = {},
        month = {April},
        }