CT Image Reconstruction Based On Non-local Low-rank Regularization

In recent years, due to the concept of compressive sensing, the medical image reconstruction problem has been a strong concern of scholars.In signal processing,sparsity has been widely exploited for exact reconstruction of a signal from a small number of random measurements. Recent advances have suggested that structured or group sparsity often leads to more powerful signal reconstruction techniques in various compressed sensing(CS) studies. It takes advantage of the natural image and medical image itself has self similarity, to construct the image blocks of similar set, using the matrix of these similar blocks that is sparse and compressed sensing theory to reconstruct the image. In some scholars’ papers, they propose a nonlocal low-rank regularization(NLR) approach toward exploiting structured sparsity and explore its application into CS of both photographic and MRI images. And they also propose the use of a nonconvex logdet(X) as a smooth surrogate function for the rank instead of the convex nuclear norm; and justify the benefit of such a strategy using extensive experiments. Because the CT images also have the characteristic of the structured sparsity, so in this paper,I explore its application into CS of CT images.Experimental results make a good performance for image recovery.