Joint Image Sharpening and Denoising by 3d Transform-domain Collaborative Filtering

In order to simultaneously sharpen image details and attenuate noise, we propose to combine the recent blockmatching and 3D Þltering (BM3D) denoising approach, based on 3D transform-domain collaborative Þltering, with alpha-rooting, a transform-domain sharpening technique. The BM3D exploits grouping of similar image blocks into 3D arrays (groups) on which collaborative Þltering (by hard-thresholding) is applied. We propose two approaches of sharpening by alpha-rooting; the Þrst applies alpharooting individually on the 2D transform spectra of each grouped block; the second applies alpha-rooting on the 3D-transform spectra of each 3D array in order to sharpen Þne image details shared by all grouped blocks and further enhance the interblock differences. The conducted experiments with the proposed method show that it can preserve and sharpen Þne image details and effectively attenuate noise.