DESIGN AND IMPLEMENTATION OF FIRE EARLY WARNING SYSTEM IN IN-BUILDING INFASTRUCTURE BASED ON DEEP LEARNING

Abstract

A fire detection system is essential for people’s safety. During the past few
years, many approaches based on smoke sensors, humidity sensors were
proposed. However, because of the limitation of the capability of sensors, those
proposed methods are infeasible for early-warning systems. The purpose of this
paper is to present a new fire early-warning system based on deep learning. The
proposed system is designed by using sensors and surveillance cameras to detect
smoke. The system is tested on our dataset and reality indoor environment.
Experiments show that the new approach is successfully applied to various
scenarios and significant for improving the accuracy of fire smoke detection.