ABSTRACT
Wildfires listed among the most devastating natural disasters, wreak havoc on both wildlife and vegetation. This endangers forest’s wealth and wipes out the entire flora and fauna regime, disrupts a region’s biodiversity, ecology, and environment, and causes risk to domestic crops and inhabitants over there. As a result, predicting wildfires in advance and prevention of these fires from propagating in order to minimize the severity of the damage became an important research issue. This work describes design of an Internet of Things (IoT)-based system that collects data using sensor technology and applies predictive analysis method to predict occurrence of forest fires. Using this system, temperature and humidity data are gathered, evaluated, and transmitted to a cloud platform; logistic regression models are developed for predicting the risk in order to alert the inhabitants of the region. In case, an unplanned forest fire occurs, this IoT system detects and alerts the control room by specifying the location using a global positioning system module facilitating appropriate action by dispatching firefighters and informing homeowners via a manually controllable buzzer. This system also ensures a secure way to protect and preserve both wildlife and human beings from threatening forest fires.
