We present a semi-automated workflow for large scale interpretation of Hyperspectral data, founded on a novel approach of mineral mapping based on a supervised dictionary learning technique. This approach exploits the complementary information from scanning electron microscopy based automated mineralogy and hyperspectral imaging techniques for estimating mineral quantities along all boreholes. We propose that it is effectively possible to propagate the mineral quantification to the entire borehole from small samples with high resolution mineralogical information strategically selected throughout the deposit. In order to apply this type of research techniques aiming at a 3D model of the alteration areas of the entire deposit based on the hyperspectral data, it is essential to have the availability of drill cores along the whole extension of the mineral deposit. Consequently, the research was focused in a study area in the Southern Spain, the Elvira deposit of the MATSA–VALORIZA mining company, where 7 km of drill core were scanned with the hyperspectral sensors. This data repository contains 24 SEM-MLA mineral maps used as training data for the Multi-scale multi-sensor data co-registration and dictionary learning algorithm.