Efficient blind compressed sensing using sparsifying transforms with convergence guarantees and application to magnetic resonance imaging

DOI10.1137/141002293zbMATH Open1343.94013arXiv1501.02923OpenAlexW2189938900MaRDI QIDQ3454501FDOQ3454501

Authors: Saiprasad Ravishankar, Yoram Bresler 

Publication date: 25 November 2015

Published in: SIAM Journal on Imaging Sciences (Search for Journal in Brave)

Abstract: Natural signals and images are well-known to be approximately sparse in transform domains such as Wavelets and DCT. This property has been heavily exploited in various applications in image processing and medical imaging. Compressed sensing exploits the sparsity of images or image patches in a transform domain or synthesis dictionary to reconstruct images from undersampled measurements. In this work, we focus on blind compressed sensing, where the underlying sparsifying transform is a priori unknown, and propose a framework to simultaneously reconstruct the underlying image as well as the sparsifying transform from highly undersampled measurements. The proposed block coordinate descent type algorithms involve highly efficient optimal updates. Importantly, we prove that although the proposed blind compressed sensing formulations are highly nonconvex, our algorithms are globally convergent (i.e., they converge from any initialization) to the set of critical points of the objectives defining the formulations. These critical points are guaranteed to be at least partial global and partial local minimizers. The exact point(s) of convergence may depend on initialization. We illustrate the usefulness of the proposed framework for magnetic resonance image reconstruction from highly undersampled k-space measurements. As compared to previous methods involving the synthesis dictionary model, our approach is much faster, while also providing promising reconstruction quality.

Full work available at URL: https://arxiv.org/abs/1501.02923

Recommendations

• Iterative thresholding compressed sensing MRI based on contourlet transform
• Convolutional Sparse Coding in Gradient Domain for MRI Reconstruction
• Nonconvex compressed sampling of natural images and applications to compressed MR imaging
• Efficient dual ADMMs for sparse compressive sensing MRI reconstruction
• Curvelet-wavelet regularized split Bregman iteration for compressed sensing

zbMATH Keywords

inverse problemscompressed sensingmedical imagingsparse representationdictionary learningmagnetic resonance imagingsparsifying transforms

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Computational methods for sparse matrices (65F50) Biomedical imaging and signal processing (92C55) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)

Cites Work

• $rm K$-SVD: An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation
• Least angle regression. (With discussion)
• Convergence of a block coordinate descent method for nondifferentiable minimization
• Atomic Decomposition by Basis Pursuit
• Variational Analysis
• Title not available (Why is that?)
• Title not available (Why is that?)
• Matching pursuits with time-frequency dictionaries
• Decoding by Linear Programming
• Stable recovery of sparse overcomplete representations in the presence of noise
• Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
• From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images
• For most large underdetermined systems of linear equations the minimal 𝓁₁‐norm solution is also the sparsest solution
• Compressed sensing
• Proximal Alternating Minimization and Projection Methods for Nonconvex Problems: An Approach Based on the Kurdyka-Łojasiewicz Inequality
• A block coordinate descent method for regularized multiconvex optimization with applications to nonnegative tensor factorization and completion
• Adaptive greedy approximations
• Online learning for matrix factorization and sparse coding
• Sparse Approximate Solutions to Linear Systems
• Analysis versus synthesis in signal priors
• Learning Multiscale Sparse Representations for Image and Video Restoration
• Subspace Pursuit for Compressive Sensing Signal Reconstruction
• Recursive Least Squares Dictionary Learning Algorithm
• Dictionary Identification—Sparse Matrix-Factorization via $\ell_1$-Minimization
• A self-referencing level-set method for image reconstruction from sparse Fourier samples
• Sparse and Redundant Modeling of Image Content Using an Image-Signature-Dictionary
• Solving quadratically constrained least squares problems using a differential-geometric approach
• A block coordinate variable metric forward-backward algorithm
• Dictionary Learning for Sparse Approximations With the Majorization Method
• Blind Compressed Sensing
• Title not available (Why is that?)
• Learning Sparsifying Transforms
• Structured overcomplete sparsifying transform learning with convergence guarantees and applications
• Sparsifying Transform Learning With Efficient Optimal Updates and Convergence Guarantees
• Learning Doubly Sparse Transforms for Images
• Efficient blind compressed sensing using sparsifying transforms with convergence guarantees and application to magnetic resonance imaging
• Learning Filter Bank Sparsifying Transforms
• Wavelet-Based Compressed Sensing Using a Gaussian Scale Mixture Model

Cited In (8)

• Iterative thresholding compressed sensing MRI based on contourlet transform
• FRIST-flipping and rotation invariant sparsifying transform learning and applications
• Semi-Blind Sparse Image Reconstruction With Application to MRFM
• Efficient blind compressed sensing using sparsifying transforms with convergence guarantees and application to magnetic resonance imaging
• Maximum likelihood estimation of regularization parameters in high-dimensional inverse problems: an empirical Bayesian approach. II: Theoretical analysis
• Efficient dual ADMMs for sparse compressive sensing MRI reconstruction
• A four directions variational method for solving image processing problems
• Data Driven Tight Frame for Compressed Sensing MRI Reconstruction via Off-the-Grid Regularization

Uses Software

• PDCO
• DLMRI-Lab

This page was built for publication: Efficient blind compressed sensing using sparsifying transforms with convergence guarantees and application to magnetic resonance imaging

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3454501)