Nonlocal Weighted Nuclear Norm Minimization based Sparse-Sampling CT Image Reconstruction

摘要

Traditional total variation (TV) based computed tomography (CT) image reconstruction methods suffer from the notorious blocky effect while the sampling rate is low. Low-rank based method is an effective way to circumvent this side effect. Normally, nuclear norm is utilized to impose the low rank constraint and its numerical computation depends on the sum of singular values. However, since larger singular values mainly deliver the structural information, treating all the singular values equally may lead to imperfect preservation of edges and textures. To deal with this problem, we propose to reconstruct CT image by explicitly exploring the nonlocal similarity in the target image with nonlocal weighted nuclear norm minimization (NOWNUNM). The experiments show that the proposed method achieved better qualitative and quantitative results than several state-of-the-art methods.