Journal of Siberian Federal University. Engineering & Technologies / Integration of Deep Learning and Wireless Sensor Networks for Accurate Fire Detection in Indoor Environment

Full text (.pdf)
Issue
Journal of Siberian Federal University. Engineering & Technologies. 2024 17 (1)
Authors
Omer A. Dheyab; Chernikov, Dmitry Yu.; Selivanov, Alexander S.
Contact information
Omer A. Dheyab: Siberian Federal University Krasnoyarsk, Russian Federation; University of Technology-Iraq Baghdad, Iraq; ; Chernikov, Dmitry Yu. : Siberian Federal University Krasnoyarsk, Russian Federation; Selivanov, Alexander S. : Siberian Federal University Krasnoyarsk, Russian Federation
Keywords
fire detection; yolov5; deep learning; wireless sensor networks; yolov5
Abstract

Systems for detecting fires are essential for protecting people and property. Still, there are a lot of problems with these systems' accuracy and the frequency of false warnings. This study uses wireless sensor networks with deep learning to improve the accuracy of real- time fire detection systems and decrease false alarms. Wi-Fi camera movies are analyzed using the YOLOv5 deep learning model. This model locates and classifies items quickly and precisely using deep learning techniques. To guarantee accurate detection, a sizable collection of fire-related data is used to train the model. When a fire occurs, users receive early warnings via WebRTC technology, and live footage of the burning location is broadcast. Using these sophisticated technologies, the efficiency of fire detection in the indoor environment can be improved, providing users with immediate and accurate alarms. Personnel and property safety is improved, and losses due to fires in the interior environment are decreased

Pages
124–135
EDN
WZHDHV
Paper at repository of SibFU
https://elib.sfu-kras.ru/handle/2311/152586

Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).