Applying artificial intelligence models for the automatic forest fire detection

Abstract

Throughout the past decade, the development of Artificial Intelligence-based devices for the automatic detection of early stage forest fires has been a growing focus. Computer Vision techniques are well-suited for this problem due to the distinctive visual characteristics of forest fires. The effectiveness of several Artificial Intelligence algorithms in a binary classification problem involving fire/ non-fire images was assessed by comparing them using a publicly available dataset. The benchmark dataset was used to both train and evaluate the models. An optimization method was employed to train the Artificial Intelligence algorithms, resulting in a higher performance than that previously achieved by studies on the same dataset.