Integrating AI With Remote Sensing For Mineral Prospectivity Mapping

Abstract

Mineral prospectivity mapping (MPM) is an essential part of mineral discovery, which has conventionally gone through a process dependent on geoscience experience and manual data interpretation. Due to the swift development of remote sensing-based technologies and artificial intelligence (AI), in particular, machine learning (ML) and deep learning (DL), the study of the mineral-rich zones has undergone a paradigm shift. The proposed method of incorporating AI in remotely provided sensory data can be used to automate and improve the precision of MPM, so this paper tries to discuss that. The approach relies on spectral, geological, and topographical data, which are satellite-derived and processed via the application of supervised machine learning algorithms, to provide the indication of the priority of potential mineral areas. The suggested system was applied on a well documented area of mineralization where the results have shown that the effectiveness of AI models to the conventional method is far superior to the traditional in terms of accuracy of the prediction and spatial generalization. The paper demonstrates the usefulness of such an integration that could be used as an aid by geologists during decision-making exercises reducing field survey expenditures and maximising the exploration expenses.