Mineral mapping
Minerals show characteristic features in different parts of the electromagnetic spectrum which provide a wealth of information about their abundance and composition in the ore. Based on this principle, we acquire hyperspectral data with sensors that cover the visible and near infrared (VNIR) range, short-wave infrared (SWIR) range, mid-wave infrared (MWIR) range and long-wave infrared (LWIR) range. The area of application of the sensors extends to different platforms (e.g. drones, camera tripods, conveyor belts).
Rapid analysis of drill-cores, geological samples and outcrops
At HIF, we deploy hyperspectral sensors to acquire data at different size scales. Hyperspectral scanners are used as a non-destructive technique to efficiently map the abundance and distribution of minerals, for example, in drill-cores and geological samples. Hyperspectral cameras are also used for the non-invasive characterization of different mineral assemblages, structural features and alteration patterns in outcrops. To this end, we apply physics-based analyses and machine learning approaches to map mineralogy at high resolution and over a large area. To fully leverage the potential of hyperspectral datasets, these techniques can be fully automated and combine information from multiple sensors in order to rapidly identify minerals of interest and deliver interpretable results.
Flowchart of the proposed methodology to map minerals at a high-spatial resolution using hyperspectral data. (I.C. Contreras Acosta et al, 2021), Source: HZDR/HIF