Fusion of wave atom-based Wiener shrinkage filter and joint non-local means filter for texture-preserving image deconvolution

Abstract

We propose an efficient texture-preserving image deconvolution algorithm. This algorithm restores a blurred image by incorporating a wave atom-based Wiener shrinkage filter with a spatial-based joint non-local means filter. Wave atom is a new transform which is half multi-scale and half multi-directional. This transform offers a better representation of images containing oscillatory patterns and textures than other known transforms. Our method first restores the image in the frequency domain to obtain a noisy result with minimal loss of image components, followed by a Wiener shrinkage filter in the wave atom domain to attenuate the leaked colored noise. Although the wave atom-based method is efficient in texture-preserving image denoising, it is prone to producing edge ringing which relates to the structure of the underlying wave atom. In order to reduce the ringing, we developed an efficient joint non-local means filter by using the wave atom deconvolution result. This filter could suppress the leaked colored noise while preserving image details. We compare our deconvolution algorithm with many competitive deconvolution techniques in terms of ISNR and visual quality.