On the decomposition of Mars hyperspectral data by ICA and Bayesian positive source separation

The surface of Mars is currently being imaged with an unprecedented combination of spectral and spatial resolution. This high resolution, and its spectral range, gives the ability to pinpoint chemical species on the surface and the atmosphere of Mars more accurately than before. The subject of this paper is to present a method to extract informations on these chemicals from hyperspectral images. A first approach, based on independent component analysis (ICA) [P. Comon, Independent component analysis, a new concept? Signal Process. 36 (3) (1994) 287-314], is able to extract artifacts and locations of CO"2 and H"2O ices. However, the main independence assumption and some basic properties (like the positivity of images and spectra) being unverified, the reliability of all the independent components (ICs) is weak. For improving the component extraction and consequently the endmember classification, a combination of spatial ICA with spectral Bayesian positive source separation (BPSS) [S. Moussaoui, D. Brie, A. Mohammad-Djafari, C. Carteret, Separation of non-negative mixture of non-negative sources using a Bayesian approach and MCMC sampling, IEEE Trans. Signal Process. 54 (11) (2006) 4133-4145] is proposed. To reduce the computational burden, the basic idea is to use spatial ICA yielding a rough classification of pixels, which allows selection of small, but relevant, number of pixels. Then, BPSS is applied for the estimation of the source spectra using the spectral mixtures provided by this reduced set of pixels. Finally, the abundances of the components are assessed on the whole pixels of the images. Results of this approach are shown and evaluated by comparison with available reference spectra.

[1]  Sylvain Douté,et al.  WAVANGLET: An Efficient Supervised Classifier for Hyperspectral Images , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[2]  A. Mohammad-Djafari A Bayesian approach to source separation , 2000, math-ph/0008025.

[3]  James V. Candy,et al.  Adaptive and Learning Systems for Signal Processing, Communications, and Control , 2006 .

[4]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[5]  Donald Geman,et al.  Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images , 1984 .

[6]  J. Cardoso,et al.  Blind beamforming for non-gaussian signals , 1993 .

[7]  C. Lawson,et al.  Solving least squares problems , 1976, Classics in applied mathematics.

[8]  Hichem Snoussi,et al.  Bayesian blind separation of generalized hyperbolic processes in noisy and underdeterminate mixtures , 2006, IEEE Transactions on Signal Processing.

[9]  J. W. Miskin,et al.  Independent Component Analysis: Ensemble Learning for blind source separation , 2001 .

[10]  Mark D. Plumbley,et al.  Polyphonic transcription by non-negative sparse coding of power spectra , 2004, ISMIR.

[11]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[12]  G. B. Smith,et al.  Preface to S. Geman and D. Geman, “Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images” , 1987 .

[13]  Christian P. Robert,et al.  Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[14]  Joseph G. Ibrahim,et al.  Monte Carlo Methods in Bayesian Computation , 2000 .

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

[16]  B. Kowalski,et al.  Multivariate curve resolution applied to spectral data from multiple runs of an industrial process , 1993 .

[17]  Daniel D. Lee,et al.  Bayesian regularization and nonnegative deconvolution for room impulse response estimation , 2006, IEEE Transactions on Signal Processing.

[18]  P. Paatero,et al.  Positive matrix factorization: A non-negative factor model with optimal utilization of error estimates of data values† , 1994 .

[19]  Ali Mohammad-Djafari,et al.  Bayesian analysis of spectral mixture data using Markov Chain Monte Carlo Methods , 2006 .

[20]  Charles L. Lawson,et al.  Solving least squares problems , 1976, Classics in applied mathematics.

[21]  Anand Rangarajan,et al.  Bayesian image reconstruction for transmission tomography using deterministic annealing , 2003, J. Electronic Imaging.

[22]  Daniel B. Rowe,et al.  Multivariate Bayesian Statistics: Models for Source Separation and Signal Unmixing , 2002 .

[23]  Athina P. Petropulu,et al.  Long-range dependence and heavy-tail modeling for teletraffic data , 2002, IEEE Signal Process. Mag..

[24]  John F. Mustard,et al.  Spectral unmixing , 2002, IEEE Signal Process. Mag..

[25]  P. Comon Independent Component Analysis , 1992 .

[26]  D. Brie,et al.  Separation of Non-Negative Mixture of Non-Negative Sources Using a Bayesian Approach and MCMC Sampling , 2006, IEEE Transactions on Signal Processing.

[27]  François Poulet,et al.  OMEGA: Observatoire pour la Minéralogie, l'Eau, les Glaces et l'Activité , 2004 .

[28]  Christian Jutten,et al.  Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture , 1991, Signal Process..

[29]  Jean-Francois Cardoso,et al.  Blind signal separation: statistical principles , 1998, Proc. IEEE.

[30]  Patrik O. Hoyer,et al.  Non-negative sparse coding , 2002, Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing.

[31]  David Brie,et al.  Non-negative source separation: range of admissible solutions and conditions for the uniqueness of the solution , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[32]  P. Deschamps,et al.  Atmospheric modeling for space measurements of ground reflectances, including bidirectional properties. , 1979, Applied optics.

[33]  Christian P. Robert,et al.  Monte Carlo Statistical Methods (Springer Texts in Statistics) , 2005 .

[34]  Ali Mansour,et al.  Blind Separation of Sources , 1999 .

[35]  Richard M. Everson,et al.  Independent Component Analysis: Principles and Practice , 2001 .

[36]  R. Bro,et al.  A fast non‐negativity‐constrained least squares algorithm , 1997 .

[37]  Andrzej Cichocki,et al.  Adaptive Blind Signal and Image Processing - Learning Algorithms and Applications , 2002 .

[38]  Andrzej Cichocki,et al.  Adaptive blind signal and image processing , 2002 .

[39]  P. Drossart,et al.  Perennial water ice identified in the south polar cap of Mars , 2004, Nature.

[40]  Stephen J. Roberts,et al.  Data decomposition using independent component analysis with prior constraints , 2003, Pattern Recognit..

[41]  Ali Mohammad-Djafari,et al.  Application of Bayesian non-negative source separation to mixture analysis in spectroscopy , 2004 .

[42]  T. Brown,et al.  A new method for spectral decomposition using a bilinear Bayesian approach. , 1999, Journal of magnetic resonance.

[43]  Pierre Comon,et al.  Blind separation of sources, part II: Problems statement , 1991, Signal Process..

[44]  Anand Rangarajan,et al.  Bayesian image reconstruction for transmission tomography using mixture model priors and deterministic annealing algorithms , 2001, SPIE Medical Imaging.

[45]  S. Douté,et al.  South Pole of Mars: Nature and composition of the icy terrains from Mars Express OMEGA observations , 2007 .

[46]  Mark D. Plumbley,et al.  Polyphonic music transcription by non-negative sparse coding of power spectra , 2004 .

[47]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[48]  Victoria Stodden,et al.  When Does Non-Negative Matrix Factorization Give a Correct Decomposition into Parts? , 2003, NIPS.