Artificial intelligence and data analytics for geosciences and remote sensing

To address the limitation of conventional statistics in dealing with hyperspectral data of satellite and airborne images, two contextual analyses are introduced in this chapter. The first case study presents the development of an artificial intelligence (AI) and data analytics algorithm capable of classifying hyperspectral data to support remote sensing and geographic information systems researchers in understanding and predicting changes in natural earth processes. The classification algorithm is based on a fuzzy approach combining a decision tree classifier with a fuzzy multiple-criteria decision analysis classifier. The second case study presents the development of an AI tool that extracts features from the hyperspectral data to transform a two-dimensional (2D) satellite and airborne picture to a pseudo-3D picture to improve complexity and produce multidirectional sun-shaded pictures and their edges. Such 3D images are useful in supporting the discovery of prospective ground for mineral exploration, extraction from the earth of precious minerals or other geological materials, usually from deposits of ore, veins, lodes, seams, reefs, or placer deposits, and overall to improve the efficiency and effectiveness of mineral exploration.