Bayesian separation of spectral sources under non-negativity and full additivity constraints

Authors:
Nicolas Dobigeon;Saïd Moussaoui;Jean-Yves Tourneret;Cédric Carteret
Affiliations:
University of Toulouse, IRIT/INP-ENSEEIHT, 2 rue Camichel, BP 7122, 31071 Toulouse cedex 7, France and University of Michigan, Department of EECS, Ann Arbor, MI 48109-2122, USA;IRCCyN/ECN - CNRS UMR 6597, 1 rue de la Noë, BP 92101, 44321 Nantes cedex 3, France;University of Toulouse, IRIT/INP-ENSEEIHT, 2 rue Camichel, BP 7122, 31071 Toulouse cedex 7, France;University of Nancy, LCPME - CNRS UMR 7564, 405 rue de Vandoeuvre, 54600 Villers-lès-Nancy, France
Venue:
Signal Processing
Year:
2009

Citing 19
Cited 2

Blind separation of sources, Part II: problems statement

Signal Processing
Independent component analysis, a new concept?

Signal Processing - Special issue on higher order statistics
Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications

Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications
Non-negative Matrix Factorization with Sparseness Constraints

The Journal of Machine Learning Research
Monte Carlo Statistical Methods (Springer Texts in Statistics)

Monte Carlo Statistical Methods (Springer Texts in Statistics)
Pattern Recognition and Machine Learning (Information Science and Statistics)

Pattern Recognition and Machine Learning (Information Science and Statistics)
Hyperspectral Data Exploitation: Theory and Applications

Hyperspectral Data Exploitation: Theory and Applications
Blind Unmixing of Linear Mixtures using a Hierarchical Bayesian Model. Application to Spectroscopic Signal Analysis

SSP '07 Proceedings of the 2007 IEEE/SP 14th Workshop on Statistical Signal Processing
Joint Segmentation of Multivariate Astronomical Time Series: Bayesian Sampling With a Hierarchical Model

IEEE Transactions on Signal Processing
Bayesian curve fitting using MCMC with applications to signalsegmentation

IEEE Transactions on Signal Processing
Joint Segmentation of Piecewise Constant Autoregressive Processes by Using a Hierarchical Model and a Bayesian Sampling Approach

IEEE Transactions on Signal Processing
Separation of Non-Negative Mixture of Non-Negative Sources Using a Bayesian Approach and MCMC Sampling

IEEE Transactions on Signal Processing
Identification and deconvolution of multichannel linearnon-Gaussian processes using higher order statistics and inverse filtercriteria

IEEE Transactions on Signal Processing
An MCMC sampling approach to estimation of nonstationary hiddenMarkov models

IEEE Transactions on Signal Processing
Semi-Supervised Linear Spectral Unmixing Using a Hierarchical Bayesian Model for Hyperspectral Imagery

IEEE Transactions on Signal Processing - Part I
Bayesian blind separation of generalized hyperbolic processes in noisy and underdeterminate mixtures

IEEE Transactions on Signal Processing
A Bayesian Approach for Blind Separation of Sparse Sources

IEEE Transactions on Audio, Speech, and Language Processing
Algorithms for nonnegative independent component analysis

IEEE Transactions on Neural Networks
A "nonnegative PCA" algorithm for independent component analysis

IEEE Transactions on Neural Networks

Joint Bayesian endmember extraction and linear unmixing for hyperspectral imagery

IEEE Transactions on Signal Processing
Unmixing of Hyperspectral Images using Bayesian Non-negative Matrix Factorization with Volume Prior

Journal of Signal Processing Systems

Quantified Score

Hi-index	0.08

Visualization

Abstract

This paper addresses the problem of separating spectral sources which are linearly mixed with unknown proportions. The main difficulty of the problem is to ensure the full additivity (sum-to-one) of the mixing coefficients and non-negativity of sources and mixing coefficients. A Bayesian estimation approach based on Gamma priors was recently proposed to handle the non-negativity constraints in a linear mixture model. However, incorporating the full additivity constraint requires further developments. This paper studies a new hierarchical Bayesian model appropriate to the non-negativity and sum-to-one constraints associated to the sources and the mixing coefficients of linear mixtures. The estimation of the unknown parameters of this model is performed using samples obtained with an appropriate Gibbs algorithm. The performance of the proposed algorithm is evaluated through simulation results conducted on synthetic mixture data. The proposed approach is also applied to the processing of multicomponent chemical mixtures resulting from Raman spectroscopy.