PROJECT TITLE :

Super resolution reconstruction based on block matching and three-dimensional filtering with sharpening

ABSTRACT:

Super resolution (SR) reconstruction is usually thought of to be an inverse problem in the way that unknown high resolution pictures are searched for giving low resolution images. Recent studies have shown that the sparsity regularisation employed in compressed sensing (CS) reconstruction improves the performance of SR reconstruction. Furthermore, below the assumption that mutually similar regions exist within a natural image, non-local (NL) estimation produces accurate estimates for given degraded pictures. The incorporation of this NL estimation in SR reconstruction has been shown to yield higher reconstructions. In this study, the authors propose the employment of block matching and 3-dimensional filtering with sharpening estimation because the regularisation constraint below the CS-primarily based SR framework. This estimation collects similar blocks and adaptively filters them by the shrinkage of the remodel coefficients. It recovers detailed structures whereas attenuating ringing artefacts. In addition, a sharpening technique employed in the estimation additionally emphasises edges. Therefore, the proposed SR algorithm searches for the answer that is similar to this enhanced estimate from among all possible solutions. The experimental results demonstrate that the proposed methodology provides high-quality SR pictures, each numerically and subjectively.