Image recovery from compressive sensing (CS) measurement data, especially noisy data has always been challenging due to its implicit ill-posed nature, thus, to seek a domain where a signal can exhibit a high degree of sparsity and to design an effective algorithm have drawn increasingly more attention. Among various sparsity-based models, structured or group sparsity often leads to more powerful signal reconstruction techniques. In this paper, we propose a novel entropy-based algorithm for CS recovery to enhance image sparsity through learning the group sparsity of residual. To reduce the residual of similar packed patches, the group sparsity of residual is described by a Laplacian scale mixture (LSM) model, therefore, each singular value of the residual of similar packed patches is modeled as a Laplacian distribution with a variable scale parameter, to exploit the benefits of high-order dependency among sparse coefficients. Due to the latent variables, the maximum a posteriori (MAP) estimation of the sparse coefficients cannot be obtained, thus, we design a loss function for expectation–maximization (EM) method based on relative entropy. In the frame of EM iteration, the sparse coefficients can be estimated with the denoising-based approximate message passing (D-AMP) algorithm. Experimental results have shown that the proposed algorithm can significantly outperform existing CS techniques for image recovery.
Deep neural networks have achieved the most outstanding performance in Compressed Sensing Magnetic Resonance Imaging (CS-MRI) reconstruction by learning the potential structures of images from a large number of training samples. However, the required data comprising hundreds of subjects are usually rare. In this paper, we remedy this problem by transferring the easy-to-get deep Gaussian denoisers trained with natural images for artifact reduction in the iterative recovery process without the use of full-sampled MRI data. To this end, we firstly train a set of deep Gaussian networks with natural images and then incorporate them into a play-and-plug framework that is built by modifying the proximal gradient algorithm with the classical momentum strategy. Furthermore, a nonlocal denoiser is employed for efficiently removing the artifacts. It is found that the momentum strategy can make the statistical distribution of artifacts approximately be Gaussian, making the Gaussian denoisers available for CS-MRI. In the experiments, we verify the rationality of the established framework and show that our method consistently outperforms state-of-the-art methods.
Compressive sensing (CS) has proven to be an efficient technique for accelerating magnetic resonance imaging (MRI) acquisition through breaking the Nyquist sampling limit. However, CS measurements are often corrupted by noise in the sensing process, which greatly reduces the quality of reconstructed images and deteriorates the performance of follow-up diagnosis tasks. In this paper, we propose a novel iterative shrinkage-thresholding (IST) method based on enhanced Laplacian-scaled shrinkage operation for robust CS-MRI reconstruction. Differing to existing nonlocal Laplacian-scaled based methods that easily cause biased estimation in the presence of external noise, we design a side information-aided Laplacian-scaled sparse representation model to adapt to spatially varying image structures. Reference information obtained by performing Block-Matching 3D (BM3D) thresholding on the noisy observation is incorporated into the Laplacian-scaled thresholding operator for enhancing the accuracy of sparse coding. Furthermore, we build connections between IST algorithm and approximate message passing (AMP) algorithm and consider an approximation of the divergence of thresholding, leading to an AMP-like iterative method. Experiments validate the effectiveness of leveraging a combination of Laplacian-scaled and BM3D thresholding, and demonstrate the superior robustness of the proposed method both quantitatively and visually as compared with state-of-the-art methods.