Clustered sparsity and separation of cartoon and texture

DOI10.1137/120874990zbMATH Open1280.42036arXiv1204.6123OpenAlexW1980929117MaRDI QIDQ2873221FDOQ2873221

Authors: Gitta Kutyniok 

Publication date: 23 January 2014

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

Abstract: Natural images are typically a composition of cartoon and texture structures. A medical image might, for instance, show a mixture of gray matter and the skull cap. One common task is to separate such an image into two single images, one containing the cartoon part and the other containing the texture part. Recently, a powerful class of algorithms using sparse approximation and $el{l}_1$ minimization has been introduced to resolve this problem, and numerous inspiring empirical results have already been obtained. In this paper we provide the first thorough theoretical study of the separation of a combination of cartoon and texture structures in a model situation using this class of algorithms. The methodology we consider expands the image in a combined dictionary consisting of a curvelet tight frame and a Gabor tight frame and minimizes the $el{l}_1$ norm on the analysis side. Sparse approximation properties then force the cartoon components into the curvelet coefficients and the texture components into the Gabor coefficients, thereby separating the image. Utilizing the fact that the coefficients are clustered geometrically, we prove that at sufficiently fine scales arbitrarily precise separation is possible. Main ingredients of our analysis are the novel notion of cluster coherence and clustered/geometric sparsity. Our analysis also provides a deep understanding on when separation is still possible.

Full work available at URL: https://arxiv.org/abs/1204.6123

Recommendations

• Image decomposition using optimally sparse representations and a variational approach
• Multiscale Texture Separation
• A low patch-rank interpretation of texture
• Simultaneous cartoon and texture inpainting
• Image decomposition via the combination of sparse representations and a variational approach

zbMATH Keywords

sparse approximationGabor framescurveletsshearlets\(\ell_1\) minimizationparabolic scaling

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Convex programming (90C25) Nontrigonometric harmonic analysis involving wavelets and other special systems (42C40) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) General harmonic expansions, frames (42C15) Numerical methods in Fourier analysis (65T99)

Cited In (9)

• Multi-component separation, inpainting and denoising with recovery guarantees
• Regularization graphs -- a unified framework for variational regularization of inverse problems
• Geometric separation by single-pass alternating thresholding
• A null space analysis of the \(\ell_1\)-synthesis method in dictionary-based compressed sensing
• Simultaneous cartoon and texture inpainting
• Theory and applications of compressed sensing
• Shearlets: from theory to deep learning
• Microlocal analysis of the geometric separation problem
• Analysis of simultaneous inpainting and geometric separation based on sparse decomposition

This page was built for publication: Clustered sparsity and separation of cartoon and texture

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2873221)