Image Processing Technique for Unmaned Motor Glider for Forest Fire Detection

Abstract

This article proposes an algorithm for detecting forest fires from RGB images obtained using an unmanned motor glider (UMG). The algorithm includes several stages related to finding and subtracting the background, highlighting the fire area in the RGB color space. The proposed algorithm has been tested on images of forest fires. The article proposes a variant of the flight of an unmanned motor glider in such a way as to continuously monitor large spaces for several hours. The results of calculations are presented, showing that the proposed method allows one to find areas of the image occupied by fire and can be used in automatic systems for monitoring forests from fires.