CS-MRI image reconstruction method based on sparse manifold joint constraint

Abstract

The invention discloses a CS-MRI image reconstruction method based on sparse manifold joint constraint, and belongs to the technical field of digital image processing. According to the method, the MRIimage reconstruction is realized by simultaneously utilizing norm constraint image sparsity and manifold regular term constraint image inter-block correlation. The method comprises the following steps: firstly, pre-reconstructing undersampled data of an MRI image by adopting a traditional method; finding a similar block set of the target block through a K-nearest neighbor method to obtain a structure group; establishing a graph model for each structure group, calculating an adjacent weight coefficient to construct a corresponding manifold regularization term, converting the manifold regularization term from a spatial domain to a coefficient domain, establishing a sparse manifold joint constraint reconstruction model, and finally solving the model by adopting an alternating direction multiplier method. According to the method, the manifold regularization term constraints are adopted to accurately describe the correlation of different degrees among the image blocks in different structure groups, a large amount of detail information is reserved in the reconstructed image, high reconstruction performance is obtained, and therefore the method can be used for medical image recovery.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and particularly relates to the realization of MRI image reconstruction by utilizing image sparseness and manifold structure characteristics, which is used for high-quality restoration of medical images. Background technique [0002] In recent years, with the development of Compressed Sensing (CS) theory, due to its ability to break the shackles of Nyquist sampling theorem and the advantages of being able to reconstruct signals well even at low sampling frequencies, this theory is being used more and more. It can be applied to signal recovery and reconstruction. Among all medical imaging, MRI imaging technology has the highest resolution of soft tissue, which can facilitate three-dimensional tracking of anatomical structures or lesions. However, MRI technology can make the patient unbearable due to its long imaging time, so the compressed sensing theory can be used to shorten th...

AI Technical Summary

Problems solved by technology

Common non-local regularization terms include nuclear norm and wavelet coefficient l 1 The norm is used to constrain the structural groups composed of similar image blocks, and set the same penalty parameters for all structural groups, but it is difficult to reflect the characteristics of different image similarities in different regions

Images