Traditional exploration techniques usually rely on extensive field work supported by geophysical ground surveying. However, this approach is limited by field accessibili-ty, financial means, area size and climate. Thus, we suggest to make use of multi-scale, multi-source remote sensing. A key component of this approach is the integra-tion of Remotely Piloted Airborne Systems (RPAS). We developed workflows and pro-cessing tools for an integrated analysis of hyperspectral, LIDAR, photogrammetric and magnetic data. We then use Machine Learning techniques to integrate the RPAS data into a must-scale scheme. The work-flow encompasses a satellite regional scouting followed by high resolution dro-ne-based mapping. The drone-borne hy-perspectral data is corrected of radiomet-ric and geometric distortions and provides the base for geological mapping. LIDAR data provides structural information even below canopy while magnetic data provi-des subsurface information. The validation of the RPAS data is performed via field spectroscopy and portable XRF. Machine Learning techniques serve as basis for ra-pid and accurate lithological mapping. The corrected and validated data quickly pro-vide relevant geological information. This approach was tested in several regions of the world successfully. RPAS data can vast-ly improve the accuracy of field mapping in future mineral exploration.
Powered by iCagenda