An integrated scheme to improve pan-sharpening visual quality of satellite images

Pan-sharpening is the process to fuse a low-resolution multispectral (MS) image with a high-resolution panchromatic (Pan) image to construct high spatial and spectral resolution MS image. In this study, a novel pan-sharpening scheme based on integration of Curvelet Transform (CT) and Bi-dimensional Empirical Mode Decomposition (BEMD) is investigated. First, input MS image decomposes into a sequence of intrinsic mode functions (IMFs) and residues using BEMD to remove large amounts of redundancies and carries their spatial and frequency components of all pixels. Second, decompose IMFs component and detail coefficient of Pan image using Curvelet Transform (CT), which are directional. Then, we use linear dependency to decide which detail coefficient of Pan should be injected into MS coefficient. Finally, we perform the inverse curvelet and inverse of BEMD to get high-resolution MS image. In experiments with IKONOS, Quick Bird and GeoEye satellite data, we demonstrated that our scheme has good spectral quality and efficiency. Spectral and spatial quality metrics in terms of SAM, RASE, RMSE, CC, ERGAS and QNR are used in our experiments. We compared our scheme with the state-of-the-art pan-sharpening techniques and found that our new scheme improved quantitative and qualitative results. (C) 2015 Production and hosting by Elsevier B.V. on behalf of Faculty of Computers and Information, Cairo University. This is an open access article under the CC BY-NC-ND license.