Microlocal analysis of the geometric separation problem
From MaRDI portal
Abstract: Image data are often composed of two or more geometrically distinct constituents; in galaxy catalogs, for instance, one sees a mixture of pointlike structures (galaxy superclusters) and curvelike structures (filaments). It would be ideal to process a single image and extract two geometrically `pure' images, each one containing features from only one of the two geometric constituents. This seems to be a seriously underdetermined problem, but recent empirical work achieved highly persuasive separations. We present a theoretical analysis showing that accurate geometric separation of point and curve singularities can be achieved by minimizing the norm of the representing coefficients in two geometrically complementary frames: wavelets and curvelets. Driving our analysis is a specific property of the ideal (but unachievable) representation where each content type is expanded in the frame best adapted to it. This ideal representation has the property that important coefficients are clustered geometrically in phase space, and that at fine scales, there is very little coherence between a cluster of elements in one frame expansion and individual elements in the complementary frame. We formally introduce notions of cluster coherence and clustered sparsity and use this machinery to show that the underdetermined systems of linear equations can be stably solved by minimization; microlocal phase space helps organize the calculations that cluster coherence requires.
Recommendations
Cites work
- scientific article; zbMATH DE number 205985 (Why is no real title available?)
- A Hardy space for Fourier integral operators
- A generalized uncertainty principle and sparse representation in pairs of bases
- Adaptive directional subdivision schemes and shearlet multiresolution analysis
- Aggregation for Gaussian regression
- An introduction to frames and Riesz bases
- Atomic decomposition by basis pursuit
- Beyond coherence: Recovering structured time-frequency representations
- Blind source separation by sparse decomposition in a signal dictionary
- Compressed sensing
- Continuous curvelet transform. I: Resolution of the wavefront set
- Continuous curvelet transform. II: Discretization and frames
- For most large underdetermined systems of equations, the minimal 𝓁1‐norm near‐solution approximates the sparsest near‐solution
- For most large underdetermined systems of linear equations the minimal 𝓁1‐norm solution is also the sparsest solution
- From Sparse Solutions of Systems of Equations to Sparse Modeling of Signals and Images
- Geometric separation by single-pass alternating thresholding
- Greed is Good: Algorithmic Results for Sparse Approximation
- Harmonic decomposition of audio signals with matching pursuit
- Image decomposition via the combination of sparse representations and a variational approach
- Matching pursuits with time-frequency dictionaries
- Microlocal analysis of the geometric separation problem
- Model Selection and Estimation in Regression with Grouped Variables
- Modeling textures with total variation minimization and oscillating patterns in image processing
- Morphological Component Analysis: An Adaptive Thresholding Strategy
- Multi-frame representations in linear inverse problems with mixed multi-constraints
- New tight frames of curvelets and optimal representations of objects with piecewise C2 singularities
- Optimally sparse representation in general (nonorthogonal) dictionaries via ℓ 1 minimization
- Resolution of the wavefront set using continuous shearlets
- Signal Recovery and the Large Sieve
- Simultaneous cartoon and texture image inpainting using morphological component analysis (MCA)
- Sparse representations in unions of bases
- Sparsity and persistence: mixed norms provide simple signal models with dependent coefficients
- Stable recovery of sparse overcomplete representations in the presence of noise
- Stable signal recovery from incomplete and inaccurate measurements
- The curvelet representation of wave propagators is optimally sparse
- Uncertainty Principles and Signal Recovery
- Uncertainty principles and ideal atomic decomposition
- Wavelets and curvelets for image deconvolution: a combined approach
Cited in
(42)- Shearlets: theory and applications
- Multi-component separation, inpainting and denoising with recovery guarantees
- Signal separation under coherent dictionaries and _p-bounded noise
- Microlocal analysis of edge flatness through directional multiscale representations
- Compactly supported shearlets are optimally sparse
- \(\alpha\)-Molecules
- Geometric separation by single-pass alternating thresholding
- Anisotropic multiscale systems on bounded domains
- Parabolic molecules
- Inertial proximal ADMM for separable multi-block convex optimizations and compressive affine phase retrieval
- Construction of compactly supported shearlet frames
- Theory and applications of compressed sensing
- Compressed data separation via \(\ell_q\)-split analysis with \(\ell_\infty\)-constraint
- Oracle inequalities for corrupted compressed sensing
- Shearlets: from theory to deep learning
- Stable restoration and separation of approximately sparse signals
- The wavelet transforms in Gelfand-Shilov spaces
- Compressed data separation via unconstrained l1-split analysis
- Phase retrieval of complex and vector-valued functions
- Compressed sensing with coherent and redundant dictionaries
- Inpainting via high-dimensional universal shearlet systems
- Microlocal analysis of the geometric separation problem
- Geometric separation of singularities using combined multiscale dictionaries
- Efficient resolution of anisotropic structures
- Image inpainting using sparse multiscale representations: image recovery performance guarantees
- Detection of boundary curves on the piecewise smooth boundary surface of three dimensional solids
- From compression to compressed sensing
- Analysis of inpainting via clustered sparsity and microlocal analysis
- Constructive approximation of high-dimensional functions with small efficient dimension with applications in uncertainty quantification
- Disjoint sparsity for signal separation and applications to hybrid inverse problems in medical imaging
- Iterative hard thresholding for compressed data separation
- Sparsity and incoherence in orthogonal matching pursuit
- Efficient analysis and detection of edges through directional multiscale representations
- The Cramer-Rao lower bound in a non-AWGN model for the affine phase retrieval
- Compressed data separation under _p bounded noise
- Convergence on thresholding-based algorithms for dictionary-sparse recovery
- Detection of singularities by discrete multiscale directional representations
- Geometric separation in \(\mathbb{R}^3\)
- Image separation using wavelets and shearlets
- Analysis of simultaneous inpainting and geometric separation based on sparse decomposition
- Regularization and numerical solution of the inverse scattering problem using shearlet frames
- Breaking the coherence barrier: a new theory for compressed sensing
This page was built for publication: Microlocal analysis of the geometric separation problem
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4902774)
