MAPEAMENTO DE CICATRIZES DE QUEIMADAS NO PANTANAL UTILIZANDO TÉCNICAS DE INTELIGÊNCIA ARTIFICIAL E MACHINE LEARNING

Abstract

A large number of fires are affecting the Pantanal, the world's largest wetland. Therefore, the objective of this work is to analyze the most used models for detecting fires in swamp areas, and what are the main technologies used for fire detection in swamp areas? Articles published between 2015 and 2022 were sought in the literature, which presented relevance to the researched topic, combining Artificial Intelligence with Geotechnologies. The study resulted in two chapters: a) The first chapter is a literary analysis that provides an overview of what has already been researched on the topic, highlighting the most searched keywords, authors, and themes, as well as the evolution of the topic over time; b) The second chapter consists of the files generated by the training scripts used in this study, including a shapefile of the Pantanal region. Analyses were conducted for two periods, 2023 and 2024, covering both dry and wet seasons, with the tools trained on similar regions to generate more information about the Pantanal.