IoT based Fire Detection and Warning System for Surveillance Applications using Deep Learning Model

Fire is a natural and destructive phenomenon that can cause significant damage to property and loss of life. Various devices, sensors, and methods have already been proposed and implemented to mitigate the effects of fire by detecting it in its early stages. Early detection helps to stop fire spread, saves lives, and mitigates the destruction of the infrastructure. Traditional sensors such as smoke, heat, flame, infrared, ultraviolet light radiation, and gas have been used for fire detection. These sensors are inefficient and suffer from low sensitivity, difficult deployment in large-scale applications, high false alarms, and detection delays. This paper proposes a Deep Learning model with an IoT system for monitoring, detecting, and Warning of fire in various places like forests, apartments, Industrial buildings, etc. A pre-trained Convolutional Neural Network namely MobileNet is used in Deep Learning as the base model. Transfer learning is used to develop a FireNet architecture for fire detection tasks. Deep Learning on IoT devices provides speed and accuracy in real-time detection.