PROJECT TITLE :

Image Denoising Via Collaborative Support-Agnostic Recovery - 2017

ABSTRACT:

During this paper, we have a tendency to propose a novel patch-based image denoising algorithm using collaborative support-agnostic sparse reconstruction. Within the proposed collaborative theme, similar patches are assumed to share the identical support taps. For sparse reconstruction, the chance of a faucet being active during a patch is computed and refined through a collaboration method with other similar patches in the similarity cluster. This provides a terribly smart patch support estimation, hence enhancing the quality of image restoration. Performance comparisons with state-of-the-art algorithms, in terms of PSNR and SSIM, demonstrate the prevalence of the proposed algorithm.