A review of machine learning in processing remote sensing data for mineral exploration

As a primary step in mineral exploration, a variety of features are mapped such as lithological units, alteration types, structures, and minerals. These features are extracted to aid decision-making in targeting ore deposits. Different types of remote sensing data including satellite optical and radar, airborne, and drone-based data make it possible to overcome problems associated with mapping these important parameters on the field. The rapid increase in the volume of remote sensing data obtained from different platforms has allowed scientists to develop advanced, innovative, and powerful data processing methodologies. Machine learning methods can help in processing a wide range of remote sensing data and in determining the relationship between the reflectance continuum and features of interest. Moreover, these methods are robust in processing spectral and ground truth measurements against noise and uncertainties. In recent years, many studies have been carried out by supplementing geological surveys with remote sensing data, and this area is now considered a hotspot in geoscience research. This paper reviews the implementation and adaptation of some popular and recently established machine learning methods for remote sensing data processing and investigates their applications for exploring different ore deposits. Lastly, the challenges and future directions in this critical interdisciplinary field are discussed.