PROJECT TITLE :

Full Scale Regression-Based Injection Coefficients for Panchromatic Sharpening - 2018

ABSTRACT:

Pansharpening is usually related to the fusion of a high spatial resolution however low spectral resolution (panchromatic) image with a high spectral resolution but low spatial resolution (multispectral) image. The calculation of injection coefficients through regression is a terribly popular and powerful approach. These coefficients are sometimes estimated at reduced resolution. During this Project, the estimation of the injection coefficients at full resolution for regression-based mostly pansharpening approaches is proposed. To the present aim, an iterative algorithm is proposed and studied. Its convergence, no matter the initial guess, is demonstrated in all the sensible cases and also the reached asymptotic price is analytically calculated. The performance is assessed both at reduced resolution and at full resolution on four data sets acquired by the IKONOS sensor and the WorldView-three sensor. The proposed full scale approach always shows the best performance with respect to the benchmark consisting of state-of-the-art pansharpening ways.