PROJECT TITLE :

Robust Single-Image Super-Resolution Based on Adaptive Edge-Preserving Smoothing Regularization - 2018

ABSTRACT:

Single-image super-resolution (SR) reconstruction via sparse illustration has recently attracted broad interest. It's known that a low-resolution (LR) image is prone to noise or blur thanks to the degradation of the observed image, which would lead to a poor SR performance. In this Project, we propose a novel strong edge-preserving smoothing SR (REPS-SR) methodology in the framework of sparse representation. An EPS regularization term is meant based on gradient-domain-guided filtering to preserve image edges and scale back noise in the reconstructed image. Furthermore, a smoothing-aware factor adaptively determined by the estimation of the noise level of LR pictures without manual interference is presented to obtain an optimal balance between the data fidelity term and also the proposed EPS regularization term. An iterative shrinkage algorithm is used to obtain the SR image results for LR pictures. The proposed adaptive smoothing-aware theme makes our technique sturdy to completely different levels of noise. Experimental results indicate that the proposed methodology will preserve image edges and scale back noise and outperforms the current state-of-the-art strategies for noisy pictures.