PROJECT TITLE :

A Variational Pansharpening Approach Based on Reproducible Kernel Hilbert Space and Heaviside Function - 2018

ABSTRACT:

Pansharpening is a vital application in remote sensing image processing. It will increase the spatial-resolution of a multispectral image by fusing it with a high spatial-resolution panchromatic image in the identical scene, that brings great favor for subsequent processing like recognition, detection, etc. During this Project, we propose endless modeling and sparse optimization based mostly method for the fusion of a panchromatic image and a multispectral image. The proposed model is mainly primarily based on reproducing kernel Hilbert space (RKHS) and approximated Heaviside function (AHF). Further, we additionally propose a Toeplitz sparse term for representing the correlation of adjacent bands. The model is convex and solved by the alternating direction methodology of multipliers that guarantees the convergence of the proposed method. Intensive experiments on many real datasets collected by totally different sensors demonstrate the effectiveness of the proposed technique as compared with many state-of-the-art pansharpening approaches.