Predicting Forest Areas Susceptible to Fire Risk Using Convolutional Neural Networks

Wildfires pose a grave danger and threat to both human health and the environment, which is why early detection of wildfires is crucial. In this study, a convolutional neural network, which is a deep learning technique for computer vision, that is capable of classifying satellite imaging of forest cover in Canada as either being prone to wildfires or not being prone to wildfires is created. This model achieved an accuracy of 95.06% and is not only accurate but also reliable and unbiased in terms of the training set and the test set. We also review an existing model for the same dataset. Furthermore, this study discusses the application of this model in the real world, its feasibility, its future scope, and strategies to improve it.