Enhanced Lithological Mapping in El-Missikat and El-Erediya Areas, Central Eastern Desert, Egypt, Leveraging Remote Sensing Techniques and Machine Learning Algorithms

Ali Shebl,Aya S. Shereif,A S Mahmoud,Á. Csámer
DOI: https://doi.org/10.1109/TGRS.2024.3471982
IF: 8.2
IEEE Transactions on Geoscience and Remote Sensing
Abstract:In the field of mineral exploration, it is strikingly evident that radioactive-bearing mineralization predominantly resides within granitic intrusions and along structural discontinuities. Consequently, the comprehensive mapping of lithological features emerges as a crucial means of accurately guiding the identification of these mineralizations. The present research is dedicated to enhancing the characterization of granitic rocks located within Egypt’s Central Eastern Desert (El-Missikat and El-Erediya regions). These areas have garnered attention due to their notably high concentrations of radioactive mineralizations, prompting the need for a more in-depth investigation. Despite the study area’s importance for potential radioactive mineral deposits, examining the geological map reveals notable challenges and inconveniences. Our research aims to address these issues using remote sensing data and machine learning algorithms (MLAs). We used image processing techniques, including false-color composites (FCCs), principal component analysis (PCA), and independent component analysis (ICA) to identify eight lithological targets and generate reference maps for the study area.The widely used support vector machine (SVM) was trained with informative image combinations, showing reasonable lithological allocations, especially when it fed with FCC 12-6-2 in RGB. Our study found that incorporating dimensionality-reduction techniques like PCA and ICA with FCCs significantly boosted accuracy by over 15%. Using Sentinel-2 imagery and SVM, we created a novel lithological map for the challenging study area, pinpointing mineralization-rich zones, particularly those linked to shear zones within granitic rocks. This map enhances progress in characterizing rock units, and we strongly advocate the use of dimensionality-reduction techniques, such as PCA and ICA, to feed MLAs. These techniques play a crucial role in producing precise, unbiased lithological maps for complex terrains, aiding in the localization of valuable mineral deposits.
Computer Science,Geology,Environmental Science
What problem does this paper attempt to address?