Applying optimum fusion method to improve lithological mapping of sedimentary rocks using sentinel-2 and ASTER satellite images

This study was set the yield a precise map of geological formations appliying satellite image fusion techniques. Geologic maps are one of the most valuable sources for understanding the geological conditions of a region used to show the spread of different rock and soil outcrops. In the present research, the fusion of ASTER and Sentinel-2 images were applied using Brovey Transform (BT), Gram-Schmidt (GS), Color Normalized (CN), Smoothing Filter-based Intensity Modulation (SFIM) and Discrete Wavelet Transform (DWT) methods to prepare the lithological map. Based on the results of the employed methods, DWT and BT methods are good in terms of providing spectral and spatial information, respectively. The results also showed that the SFIM method has appropriate spectral and spatial accuracy. At the classification stage, all fused images were processed through supervised classification algorithm of Support Vector Machine (SVM) to identify and separate the geological formations of the study area. The evaluation of the classifications results demonstrated that the SVM method used in the fused images by SFIM method, with 83.16 overall accuracy and 0.82 kappa coefficient has evidential results for lithological mapping.