Fire accidents pose a serious threat to human life, property, infrastructure, and the environment, making early and reliable detection essential for effective emergency response. Conventional fire detection systems primarily depend on smoke, heat, and temperature sensors, which may experience limitations in detecting visible flames at an early stage and can generate false alarms under varying environmental conditions. This research proposes an intelligent realtime fire and smoke detection system using the YOLOv8 (You Only Look Once Version 8) deep learning algorithm for automated visual surveillance. The proposed framework utilizes the Home Fire Dataset, containing diverse fire and smoke images captured under varying lighting conditions, object scales, backgrounds, and indoor environments. The dataset undergoes preprocessing, annotation verification, augmentation, training, and validation to improve model robustness and generalization. The YOLOv8 model is trained using GPU acceleration in Google Colab to simultaneously detect and localize fire and smoke regions through bounding boxes and confidence scores. Performance is evaluated using Precision, Recall, F1-score, mAP@50, mAP@50:95, and inference speed to assess detection accuracy, localization capability, and realtime performance. The proposed system aims to minimize false detections and missed hazards while enabling rapid identification of fire-related threats. Its vision-based architecture can be integrated with CCTV surveillance, IoT-enabled alarm systems, emergency notification platforms, and smart-building technologies. The framework has potential applications in residential buildings, educational institutions, hospitals, offices, warehouses, factories, hotels, commercial complexes, and other indoor environments requiring continuous monitoring. By combining deep learning and real-time computer vision, the proposed approach provides a scalable and cost-effective foundation for intelligent fire safety, supporting early hazard identification, faster emergency intervention, improved situational awareness, and safer living and working environments.

Neepa Christy N, Sanjana A N, Rishwanth S V, S Shobika, Dr Pradeepa K,"Real-Time Fire and Smoke Detection Using YOLOv8 for Intelligent Indoor Surveillance" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(4): Page(497-516) July - August 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.