Extending NMF to blindly separate linear-quadratic mixtures of uncorrelated sources

This paper proposes a new constrained method, based on nonnegative matrix factorization, for blindly separating linear-quadratic (LQ) mixtures of mutually uncorrelated source signals when the sources and mixing parameters are all nonnegative. The uncorrelatedness of the sources is used as a regularization term in the cost function. The main advantage of exploiting uncorrelatedness in this manner is that the inversion of the mixing model, which is a difficult task in the case of determined LQ mixtures, is not required, contrary to the classical LQ methods based on independent component analysis. Experimental results using artificial data and real-world chemical data confirm the effectiveness of our method.

[1]  Christian Jutten,et al.  A blind source separation method for chemical sensor arrays based on a second order mixing model , 2015, 2015 23rd European Signal Processing Conference (EUSIPCO).

[2]  Yannick Deville,et al.  Recurrent networks for separating extractable-target nonlinear mixtures. Part I: Non-blind configurations , 2009, Signal Process..

[3]  M. Castella,et al.  Inversion of Polynomial Systems and Separation of Nonlinear Mixtures of Finite-Alphabet Sources , 2008, IEEE Transactions on Signal Processing.

[4]  Yannick Deville,et al.  Blind Separation of Linear-Quadratic Mixtures of Real Sources Using a Recurrent Structure , 2009, IWANN.

[5]  Yannick Deville,et al.  Linear–Quadratic Mixing Model for Reflectances in Urban Environments , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Pierre Comon,et al.  Handbook of Blind Source Separation: Independent Component Analysis and Applications , 2010 .

[7]  Farnood Merrikh-Bayat,et al.  A nonlinear blind source separation solution for removing the show-through effect in the scanned documents , 2008, 2008 16th European Signal Processing Conference.

[8]  Yannick Deville,et al.  Recurrent networks for separating extractable-target nonlinear mixtures. Part II. Blind configurations , 2013, Signal Process..

[9]  Guillermo Bedoya Jimenez.  Non-linear blind signal separation for chemical solid-state sensor arrays , 2006 .

[10]  Mark D. Plumbley.  Conditions for nonnegative independent component analysis , 2002, IEEE Signal Processing Letters.

[11]  Christian Jutten,et al.  Application of Blind Source Separation Methods to Ion-Selective Electrode Arrays in Flow-Injection Analysis , 2014, IEEE Sensors Journal.

[12]  Yannick Deville,et al.  Linear-Quadratic Blind Source Separation Using NMF to Unmix Urban Hyperspectral Images , 2014, IEEE Transactions on Signal Processing.

[13]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[14]  Luís B. Almeida,et al.  Nonlinear separation of show-through image mixtures using a physical model trained with ICA , 2012, Signal Process..

[15]  W. Verstraeten,et al.  Nonlinear Hyperspectral Mixture Analysis for tree cover estimates in orchards , 2009 .

[16]  Nikos D. Sidiropoulos,et al.  Non-Negative Matrix Factorization Revisited: Uniqueness and Algorithm for Symmetric Decomposition , 2014, IEEE Transactions on Signal Processing.

[17]  Messaoud Benidir,et al.  Blind identification of a linear-quadratic model using higher-order statistics , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[18]  Karim Abed-Meraim,et al.  Blind identification of a linear-quadratic mixture of independent components based on joint diagonalization procedure , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.

[19]  Yannick Deville,et al.  Blind Separation of Parametric Nonlinear Mixtures of Possibly Autocorrelated and Non-Stationary Sources , 2014, IEEE Transactions on Signal Processing.

[20]  Leonardo Tomazeli Duarte.  Design of Smart Chemical Sensor Arrays: an Approach Based on Source Separation Methods , 2009 .

[21]  Zhaoshui He,et al.  Extended SMART Algorithms for Non-negative Matrix Factorization , 2006, ICAISC.

[22]  K. Wilson,et al.  SPECTROGRAM DIMENSIONALITY REDUCTION WITH INDEPENDENCE CONSTRAINTS , 2010 .