Enhancing Fire Safety: A CCTV-based Alert System

• Published in: 2024 5th International Conference on Smart Electronics and Communication (ICOSEC) pp. 129 - 134
• Main Authors: Harish, G., Varshitha, V., Kumar, K.V.V., Suneel, E., Valluri, Aswini, SandhyaRani, P.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 18.09.2024
• Subjects: Cameras; Closed Circuit Television Surveillance (CCTV); Email; Fire; Fire safety; Image processing techniques; Machine learning; Protection; Sensor systems; Streaming media; Streams; Surveillance; YOLO; You Only Look Once Version3
• DOI: 10.1109/ICOSEC61587.2024.10722728

Enhancing the protection and safety of people and property in a variety of situations and environments requires the critical duty of fire detection. Here, a fire alert system that monitors the video streams of a specific region using CCTV surveillance cameras to identify fires was proposed. The system processes and analyses the video frames to find patterns and features related to fire in the provided image. It does this by using machine learning methods and image processing techniques. The location and fire-detected photo are sent together with an email alert that the system generates and sends to the user or the appropriate agency. The system can scale and adjust to various environmental conditions. The camera records live video and functions as a CCTV surveillance system feed from the area under observation. The video capture module divides the video stream into separate frames and preprocesses each one using methods like edge detection. The YOLOV3 machine algorithm, a machine learning method, is used by the fire detection module to process each frame and detect fires.