Satellite Multispectral Image Enhancement based on Pan-sharpening using NSC-Transform

Abstract: In remote sensing, images acquired by various earth observation satellites tend to have either a high spatial and low spectral resolution or vice versa. Pan-sharpening is a technique which aims to improve spatial resolution of multispectral image. The challenges involve in the pan-sharpening are not only to improve the spatial resolution but also to preserve spectral quality of the multispectral image. Pan-sharpening for satellite Panchromatic (PAN) and Multispectral (MS) images involving Non-sub sampled Contourlet Transform is considered in this work. NSCT approaches with different levels of decomposition and up sampling done using Gabor based fusion. NSC-Transform is very efficient in representing the directional information and capturing intrinsic geometrical structures of the objects.In the propose method, a given number of decomposition levels are used for multispectral (MS) images and higher number of decomposition levels are used for Pan images. This decreases computation time and preserves both spectral and spatial qualities. Up sampling of MS images is performed after NSCT and not before. By applying upsampling after NSCT, structures and detail information of the MS images are more given be preserved and thus stay more distinguishable. Hence, we propose to exploit this property in pan-sharpening by fusing it with detail information provided every Pan image at the same fine level. The proposed method is tested on WorldView-2 datasets. Both spectral and spatial qualities have been improved.

Keywords: Pan-sharpening, Non-sub sampled Contourlet Transform, Quality Assessment etc.