Successive nonnegative projection algorithm for robust nonnegative blind source separation

DOI10.1137/130946782MaRDI QIDQ2876848zbMATH OpenOpenAlexFDO

Authors Nicolas Gillis

Publication date 20 August 2014

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

Full work available at URL https://arxiv.org/abs/1310.7529

zbMATH Keywords

algorithmhyperspectral unmixingnonnegative blind source separationnonnegative matrix factorizationpure-pixel assumptionrobustness to noiseseparability

Mathematics Subject Classification ID

Factorization of matrices (15A23) Positive matrices and their generalizations; cones of matrices (15B48)

Abstract: In this paper, we propose a new fast and robust recursive algorithm for near-separable nonnegative matrix factorization, a particular nonnegative blind source separation problem. This algorithm, which we refer to as the successive nonnegative projection algorithm (SNPA), is closely related to the popular successive projection algorithm (SPA), but takes advantage of the nonnegativity constraint in the decomposition. We prove that SNPA is more robust than SPA and can be applied to a broader class of nonnegative matrices. This is illustrated on some synthetic data sets, and on a real-world hyperspectral image.

Recommendations

• Blind separation of non-negative source signals using multiplicative updates and subspace projection
• A Convex Analysis Framework for Blind Separation of Non-Negative Sources
• Non‐negative matrix factorization and its application in blind sparse source separation with less sensors than sources
• Blind Source Separation in Nonlinear Mixtures: Separability and a Basic Algorithm
• Provably robust blind source separation of linear-quadratic near-separable mixtures
• Non-Negative Blind Source Separation Algorithm Based on Minimum Aperture Simplicial Cone
• Nonnegative mixture for underdetermined blind source separation based on a tensor algorithm
• Underdetermined sparse blind source separation of nonnegative and partially overlapped data

Cited in

(20)

• Efficient preconditioning for noisy separable nonnegative matrix factorization problems by successive projection based low-rank approximations
• Bayesian analysis of dynamic linear topic models
• Blind separation of non-negative source signals using multiplicative updates and subspace projection
• Convex programming based spectral clustering
• Robustness analysis of preconditioned successive projection algorithm for general form of separable NMF problem
• A survey on deep matrix factorizations
• Enhancing pure-pixel identification performance via preconditioning
• Non-Negative Blind Source Separation Algorithm Based on Minimum Aperture Simplicial Cone
• Simplex-Structured Matrix Factorization: Sparsity-Based Identifiability and Provably Correct Algorithms
• Provably robust blind source separation of linear-quadratic near-separable mixtures
• Refinement of Hottopixx method for nonnegative matrix factorization under noisy separability
• Coseparable Nonnegative Matrix Factorization
• Matrix-wise _0-constrained sparse nonnegative least squares
• A Blind Source Separation Technique for Document Restoration
• Smoothed separable nonnegative matrix factorization
• PD-ML-Lite: Private Distributed Machine Learning from Lightweight Cryptography
• Coseparable nonnegative tensor factorization with t-CUR decomposition
• On the robustness of the successive projection algorithm
• Dual simplex volume maximization for simplex-structured matrix factorization
• Rethinking coupled tensor analysis for hyperspectral superresolution: recoverable modeling under endmember variability

This page was built for publication: Successive nonnegative projection algorithm for robust nonnegative blind source separation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2876848)