Pansharpening for Better Spectral and Spatial Clarity

Abstract

Pansharpening is the fusion of a finer spectral resolution of a multispectral (MS) image with a coarser spatial resolution of a panchromatic (PAN) image. Conventional pan-sharpening algorithms can be addressed by utilizing band-dependent spatial-detail (BDSD). A filter-based approach finds a prominent application for retaining both the spatial information on the spectral quality. A Pan sharpening algorithm is proposed that uses the optimal filtered PAN image and merges with the MS image to achieve a superior Band-Dependent Spatial-Detail approach. The optimal filter injection in the BDSD approach extracts from the image the necessary and non-redundant data, which makes the BDSD approach an optimal solution for more spectral and spatial details. As compared to the conventional optimal filter and BDSD techniques, Qualitative and Quantitative analysis show that the proposed algorithm improves the fusion quality in terms of Peak Signal to Noise Ratio, Structural Similarity Index, Relative dimensionless global error in synthesis, spectral angle mapper, universal image quality index with the introduction of two parameters- quality without reference Entropy and Blind/ Referenceless Image Spatial Quality Evaluator -BRISQUE.