Super-Resolution Image Reconstruction Based on the Minimal Surface Regularization

The super-resolution image reconstruction is an ill-posed problem, which need regularizing during the reconstruction. A new regularization algorithm is presented, which regards the super-resolution image as a surface in the three-dimensional Euclidean space, and the regularization constraint is that the reconstruct image has the minimal surface area. An energy functional based on the super-resolution image reconstruction model and the regularized minimal surface is drawn, and a partial differential equation presented via the calculus of variations. The equation is solved by an explicit iteration method. The experiment shows this algorithm could reconstruct the super-resolution image efficiently.