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{192558,
        author = {Sarthak Patil and Shulmit Sasane and Prasad Wagh and Somesh Kalaskar},
        title = {AI-Powered Real-Time Fire Detection Using Deep Learning and Computer Vision},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {4729-4732},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=192558},
        abstract = {Fire incidents pose a significant threat to human life, infrastructure, and natural ecosystems. Traditional fire detection systems based on smoke and temperature sensors often suffer from delayed response, limited coverage, and high false alarm rates. This paper presents an AI-powered real-time fire detection system that leverages deep learning and computer vision techniques to identify fire events from live video streams. A lightweight convolutional neural network based on MobileNetV2 is trained using fire and non-fire image datasets. The system performs real-time inference, computes a fusion-based fire con- fidence score using visual and color-based cues, and triggers alerts through alarms, notifications, and a web-based dashboard. Experimental results demonstrate high accuracy, robustness, and low latency, making the proposed system suitable for continuous surveillance and early fire warning applications. The proposed system achieves an average inference speed of approximately 18– 22 frames per second on a standard CPU-based system.},
        keywords = {Fire Detection, Deep Learning, Computer Vi- sion, CNN, MobileNetV2, Real-Time Surveillance},
        month = {February},
        }