Image fusion based on TV -L1-convex constrained algorithm

Abstract

This paper solves the fusion problem by TV - L1-convex constrained algorithm. The energy function mainly consists of three components. One part ensures injection of more correlated detailed spatial information using total variation method. Another part preserves the spectral information through L1 norm based on data fitting term. The third part is consisted of three convex constrained conditions, which ensure that the result is proper. Because the energy function is non-smooth, we use the split Bregman algorithm to solve it. Experimental result demonstrates the superiority of the proposed method over some classical methods.