PROJECT TITLE :

A Fast Superresolution Image Reconstruction Algorithm

ABSTRACT:

During a previous paper we tend to have proposed 2 new superresolution image reconstruction algorithms, based mostly on a non-parametric numerical integration Bayesian inference technique, the Integrated Nested Laplace Approximation (INLA). Despite achieving superior image reconstruction results compared to different state-of-the-art methods, such algorithms manipulate huge matrices (though sparse). Therefore, the demand for memory usage and computation is high. During this paper, review such algorithms, solving these issues through relaxing one equation in the original mathematical model and involving the high-resolution (HR) image in a very Torus. The result's a meaningful reduction within the computation value of such algorithms and in the size of the matrices handled as well (from n2-by-n2 to n-by-n, the dimensions of the HR image). The result's a new algorithm, abundant faster than its previous version and other meaningful state-of-the-art algorithms.