ABSTRACT
Forest fire is one of the worst devastating events in the environment. It can harm the lives of plants and animals. This in turn results in climate change, the number of forest fires increased drastically in recent decades. Early-stage detection can reduce the impact of catastrophic fire. The Identification and detection of forest fire in the initial stage is an extremely difficult task even today. The satellite monitoring system is widely used in the detection of a forest fire, but it can only detect fire after a long time. The use of Unmanned Aerial vehicles for surveillance has increased tremendously in recent years. The delay in detection can be minimized by using Unmanned Aerial Vehicle and Wireless Sensor Network. In this paper, we wish to introduce a new approach for the detection of a forest fire. The aerial vehicle equipped with a camera will capture the image of fire, machine learning algorithms are used to predict the output. Wireless Sensor Network will increase the accuracy of prediction by combining the data from the sensor network. The key in this system is easy decision-making by analyzing the image from the Unmanned Aerial Vehicle within a short time.
