An Effective Super-Resolution Reconstruction Method for Geometrically Deformed Image Sequences

Despite of the technology advancements, remote sensing images usually suffer from a poor spatial resolution. To resolve this issue, a lot of research efforts have been devoted to developing resolution enhancement methods which retrieve a high-resolution image out of its low-resolution degraded versions. In this paper, we consider a nonlocal total variation (NLTV) based super-resolution method which handles low-resolution images with geometric deformations. In particular, we apply the framework of alternating direction method of multipliers (ADMM) to deduce an effective algorithm, which involves soft thresholding and gradient descent. Effectiveness and robustness to noise of the proposed method are verified by various numerical experiments.