A Review On Forest Fire Prediction Techniques

Authors:

Abstract
Forest fires are fires caused in the forest due to environmental condition like dryness if continued for a very long time can cause it to become massive which is very difficult to control. Wildfires can erupt on their own due to lightning or the heat of the sun. Unsupervised campfires, abandoned cigarettes, and other human impacts are furthermore responsible for the majority of wildfires. Techniques studied in this paper are-Deep Learning, Machine Learning, and Artificial Intelligence, and Image Processing. Under these techniques, various algorithms like Genetic Algorithm, Fuzzy Logic, Decision Tree, Multilayer Perceptron (MLP), Artificial Neural Networks (ANN), Random Forest (RF), Spatial Wavelet Analysis for Colour Variations were studied. When compared to alternative techniques, the Artificial Intelligence approach produces better outcomes. The average accuracy obtained was higher than 90%.

This publication has 10 references indexed in Scilit: