Multiple Multi-Spectral Remote Sensing Data Fusion and Integration for Geological Mapping

This paper investigates spaceborne multiple multispectral data-fusion and blending to generate an integrated data with higher spatio-spectral resolution and spectral coverage in order to obtain improved geological mapping. A hybrid approach using Gram-Schmidt pan-sharpening and Inverse Distance Weighting (IDW) based downsampling technique is developed to generate integrated data from multiple multispectral data. In this study, Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), Landsat 8, and Sentinel-2 data have been used to evaluate the developed approach for lithological mapping. Liikavaara to Puoltikasvaara including Nautanen and nearby-mining area, in the Gallivare district of Norrbotten county, Sweden, is chosen as a case study. Lithological map of the study area is produced using Support Vector Machine (SVM) classifier. Bedrock geological map from the Geological Survey of Sweden (SGU) is used for classification accuracy assessment. The results show that integrated data produced better accuracy than original individual spaceborne multispectral data for lithological mapping of the study area.