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{184561,
        author = {Annam Archita and S Zahoor Ul Haq},
        title = {A Vision-Based Traffic Accident Detection System Using a DenseNet Model for Smart City Infrastructure},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {4},
        pages = {2005-2013},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=184561},
        abstract = {In today’s smart city traffic, detecting accidents quickly is very important for safety and smooth traffic management. In this project, we present a vision-based accident detection system that works in real time using camera feeds. The system uses RGB frames along with optical flow and applies a lightweight CNN model to detect accidents. Our approach mainly focuses on solving problems like less training data and imbalanced datasets, while keeping the model simple and fast enough for practical use. The design is cost-effective and can be deployed on smart city infrastructures like roadside cameras and IoT devices.},
        keywords = {CNN, computer vision, deep learning, smart cities, and traffic accident detection},
        month = {September},
        }