GRSG 34th Conference 2023

Title: Optimizing Lithological Mapping and Mineral Potential Assessment through the Fusion of Multi-Sensor Remote Sensing Images, Airborne Radiometric Data, and Machine Learning Algorithms

Author: Ali Shebl

Abstract:

Multi-aspect datasets hold significant promise for enhancing lithological mapping—an essential foundation for targeting mineral-rich regions: This study focuses on the integration of an extensive array of Earth observation data sources, including Sentinel 2, ASTER, and ALOS PALSAR DEM, to bolster lithological mapping within Egypt's central Eastern Desert.

To achieve lithological accurate classification, two classifiers were applied: Maximum Likelihood Classifier (MLC), and Support Vector Machine (SVM). The resulting thematic maps served as a crucial groundwork for identifying hydrothermal alterations, a process further refined by incorporating airborne magnetic and radiometric geophysical data. Notably, SVM exhibited superior performance in lithological mapping, with the Digital Elevation Model (DEM) emerging as a pivotal factor in achieving precise allocation.

Significantly, the coincidence of heightened magnetic anomalies, elevated K/eTh ratios, and detected alterations through remote sensing data substantiated the presence of real alteration anomalies. To validate the findings obtained through remote sensing and airborne geophysical indicators, extensive fieldwork and petrographic investigations were meticulously carried out .

Multi-aspect datasets hold significant promise for enhancing lithological mapping—an essential foundation for targeting mineral-rich regions: This study focuses on the integration of an extensive array of Earth observation data sources, including Sentinel 2, ASTER, and ALOS PALSAR DEM, to bolster lithological mapping within Egypt's central Eastern Desert.

To achieve lithological accurate classification, two classifiers were applied: Maximum Likelihood Classifier (MLC), and Support Vector Machine (SVM). The resulting thematic maps served as a crucial groundwork for identifying hydrothermal alterations, a process further refined by incorporating airborne magnetic and radiometric geophysical data. Notably, SVM exhibited superior performance in lithological mapping, with the Digital Elevation Model (DEM) emerging as a pivotal factor in achieving precise allocation. Significantly, the coincidence of heightened magnetic anomalies, elevated K/eTh ratios, and detected alterations through remote sensing data substantiated the presence of real alteration anomalies. To validate the findings obtained through remote sensing and airborne geophysical indicators, extensive fieldwork and petrographic investigations were meticulously carried out