Quaternion Non-negative Matrix Factorization: definition, uniqueness and algorithm

Abstract

This article introduces quaternion non-negative matrix factorization (QNMF), which generalizes the usual non-negative matrix factorization (NMF) to the case of polarized signals. Polarization information is represented by Stokes parameters, a set of 4 energetic parameters widely used in polarimetric imaging. QNMF relies on two key ingredients: (i) the algebraic representation of Stokes parameters thanks to quaternions and (ii) the exploitation of physical constraints on Stokes parameters. These constraints generalize non-negativity to the case of polarized signals, encoding positive semi-definiteness of the covariance matrix associated which each source. Uniqueness conditions for QNMF are presented. Remarkably, they encompass known sufficient uniqueness conditions from NMF. Meanwhile, QNMF further relaxes NMF uniqueness conditions requiring sources to exhibit a certain zero-pattern, by leveraging the complete polarization information. We introduce a simple yet efficient algorithm called quaternion alternating least squares (QALS) to solve the QNMF problem in practice. Closed-form quaternion updates are obtained using the recently introduced generalized HR calculus. Numerical experiments on synthetic data demonstrate the relevance of the approach. QNMF defines a promising generic low-rank approximation tool to handle polarization, notably for blind source separation problems arising in imaging applications.

arXiv comment14 pages, 2 figures

References in corpus (11)

• The Complex Gradient Operator and the CR-Calculus
• Nonnegative Matrix Factorization for Signal and Data Analytics: Identifiability, Algorithms, and Applications
• Quiet Sun internetwork magnetic fields from the inversion of Hinode measurements
• Quaternion Gradient and Hessian
• The Theory of Quaternion Matrix Derivatives
• Quaternion Derivatives: The GHR Calculus
• Identifiability of the Simplex Volume Minimization Criterion for Blind Hyperspectral Unmixing: The No Pure-Pixel Case
• On Identifiability of Nonnegative Matrix Factorization
• The Low-Frequency Solar Corona in Circular Polarization
• Spectral analysis of stationary random bivariate signals
• A complete framework for linear filtering of bivariate signals

Cited by in corpus (3)

• A general framework for constrained convex quaternion optimization
• Convex Quaternion Optimization for Signal Processing: Theory and Applications
• Blind source separation in 3rd generation gravitational-wave detectors