Data-driven thresholding in denoising with spectral graph wavelet transform

DOI10.48550/ARXIV.1906.01882zbMATH Open1458.94086arXiv1906.01882OpenAlexW3113715080MaRDI QIDQ76309FDOQ76309

Baptiste Olivier, Basile de Loynes, Fabien Navarro, Baptiste Olivier, Basile de Loynes, Fabien Navarro

Publication date: 5 June 2019

Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)

Abstract: This paper is devoted to adaptive signal denoising in the context of Graph Signal Processing (GSP) using Spectral Graph Wavelet Transform (SGWT). This issue is addressed emph{via} a data-driven thresholding process in the transformed domain by optimizing the parameters in the sense of the Mean Square Error (MSE) using the Stein's Unbiased Risk Estimator (SURE). The SGWT considered is built upon a partition of unity making the transform semi-orthogonal so that the optimization can be performed in the transformed domain. However, since the SGWT is over-complete, the divergence term in the SURE needs to be computed in the context of correlated noise. Two thresholding strategies called coordinatewise and block thresholding process are investigated. For each of them, the SURE is derived for a whole family of elementary thresholding functions among which the soft threshold and the James-Stein threshold. This multi-scales analysis shows better performance than the most recent methods from the literature. That is illustrated numerically for a series of signals on different graphs.

Full work available at URL: https://arxiv.org/abs/1906.01882

Recommendations

• Hysteresis thresholding: a graph-based wavelet block denoising algorithm
• Fractional spectral graph wavelets and their applications
• Wavelets on graphs via spectral graph theory
• scientific article
• Sparse representation on graphs by tight wavelet frames and applications

zbMATH Keywords

denoisingspectral graph theoryspectral graph wavelet transformStein unbiased risk estimationtight framevariance estimation

Mathematics Subject Classification ID

Graphs and linear algebra (matrices, eigenvalues, etc.) (05C50) Nontrigonometric harmonic analysis involving wavelets and other special systems (42C40) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Cites Work

• Estimation of the mean of a multivariate normal distribution
• Siegel's formula via Stein's identities
• Adaptive wavelet estimation: A block thresholding and oracle inequality approach
• Recurrence of distributional limits of finite planar graphs
• Local behavior of sparse analysis regularization: applications to risk estimation
• Diffusion maps
• Adapting to Unknown Smoothness via Wavelet Shrinkage
• Title not available (Why is that?)
• A wavelet tour of signal processing. The sparse way.
• The DFS Fused Lasso: Linear-Time Denoising over General Graphs
• A SURE Approach for Digital Signal/Image Deconvolution Problems
• Tight Wavelet Frames on Multislice Graphs
• Graph Wavelets for Multiscale Community Mining
• Spectrum-Adapted Tight Graph Wavelet and Vertex-Frequency Frames
• Global Image Denoising
• Signal-Adapted Tight Frames on Graphs
• Title not available (Why is that?)
• Learning Theory and Kernel Machines
• Title not available (Why is that?)
• Distribution of the Ratio of the Mean Square Successive Difference to the Variance
• Wavelets on graphs via spectral graph theory
• The solution path of the generalized lasso
• Multiscale representation of surfaces by tight wavelet frames with applications to denoising
• On total variation minimization and surface evolution using parametric maximum flows
• Towards a theoretical foundation for Laplacian-based manifold methods

Cited In (3)

• Title not available (Why is that?)
• Localized Fourier analysis for graph signal processing
• gasper

This page was built for publication: Data-driven thresholding in denoising with spectral graph wavelet transform

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q76309)