Aplicações geológicas de sensoriamento remoto na pesquisa do potencial mineral para ferro

Abstract

The state of Bahia is one of the largest producers of mineral resources in Brazil, thanks to the geodiversity of its territory, which enables the extraction of a wide variety of mineral types. However, despite this prominence, the mining-industrial production sector for iron does not contribute significantly to this prominent position, even though it has potentially minable deposits, known through countless geological surveys carried out in the last 80 years. The scientific knowledge is dispersed in the means of publication, while quantitative and qualitative data about the iron deposits are unavailable, due to the strategic value for the companies that have invested their resources to obtain them, and a small part is made available by the public organs of mineral research. Within this scenario, strategies such as bibliometry in the analysis and systematization of scientific studies on the use of remote sensing in ore evaluation may be promising in terms of producing a crucial and systematic analysis of the literature and providing subsidies that lead to the construction of information on the mapping of iron ore deposits. And, new integration methods, such as machine learning algorithms, have the potential to achieve greater accuracy in delineating areas of mining interest. Thus, this work aims to contribute to the mapping of ferrous mineral occurrences in the state of Bahia. The study is divided into two chapters. The first chapter consists of a bibliometric analysis, through text mining techniques, to recognize patterns in the literature that can serve as a basis for statistically analyzing the results of different studies on the mapping of ferrous deposits, to evaluate methodological trends in this field of action. The second chapter deals with the application of a machine learning method to remote sensing images. The use of auxiliary data, such as mineral occurrence points, geology and mining requirement areas were used to optimize the management of the results and allowing their effective use as a guideline in the mineral research process.