A variational Bayesian approach for unsupervised super-resolution using mixture models of point and smooth sources applied to astrophysical map-making

We present, in this paper, a new unsupervised method for joint image super-resolution and separation between smooth and point sources. For this purpose, we propose a Bayesian approach with a Markovian model for the smooth part and Student's t-distribution for point sources. All model and noise parameters are considered unknown and should be estimated jointly with images. However, joint estimators (joint MAP or posterior mean) are intractable and an approximation is needed. Therefore, a new gradient-like variational Bayesian method is applied to approximate the true posterior by a free-form separable distribution. A parametric form is obtained by approximating marginals but with form parameters that are mutually dependent. Their optimal values are achieved by iterating them till convergence. The method was tested by the model-generated data and a real dataset from the Herschel space observatory.

[1]  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.

[2]  Yves Goussard,et al.  Recursive deconvolution of Bernoulli-Gaussian processes using a MA representation , 1989 .

[3]  Bruno Torrésani,et al.  Random Models for Sparse Signals Expansion on Unions of Bases With Application to Audio Signals , 2008, IEEE Transactions on Signal Processing.

[4]  Pramod K. Varshney,et al.  Super-resolution land cover mapping using a Markov random field based approach , 2005 .

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

[6]  V. Šmídl,et al.  The Variational Bayes Method in Signal Processing , 2005 .

[7]  A. Gorokhov,et al.  Subspace-based techniques for blind separation of convolutive mixtures with temporally correlated sources , 1997 .

[8]  G. Wergen,et al.  Records in stochastic processes—theory and applications , 2012, 1211.6005.

[9]  M. Halpern,et al.  SANEPIC: A Mapmaking Method for Time Stream Data from Large Arrays , 2007, 0711.3462.

[10]  Michel Defrise,et al.  Inverse imaging with mixed penalties , 2004 .

[11]  Hichem Snoussi,et al.  Bayesian Unsupervised Learning for Source Separation with Mixture of Gaussians Prior , 2004, J. VLSI Signal Process..

[12]  Albert Bijaoui,et al.  Blind source separation and analysis of multispectral astronomical images , 2000 .

[13]  Thierry Chonavel,et al.  Estimation of Multipath Channels With Long Impulse Response at Low SNR via an MCMC Method , 2007, IEEE Transactions on Signal Processing.

[14]  Mohamed-Jalal Fadili,et al.  Morphological Component Analysis: An Adaptive Thresholding Strategy , 2007, IEEE Transactions on Image Processing.

[15]  T. Rodet,et al.  A gradient-like variational Bayesian algorithm , 2011, 2011 IEEE Statistical Signal Processing Workshop (SSP).

[16]  R. Stompor,et al.  MADmap: A MASSIVELY PARALLEL MAXIMUM LIKELIHOOD COSMIC MICROWAVE BACKGROUND MAP-MAKER , 2009, 0906.1775.

[17]  Martin J. Wainwright,et al.  Scale Mixtures of Gaussians and the Statistics of Natural Images , 1999, NIPS.

[18]  Kevin M. Cuomo,et al.  Super-resolution methods for wideband radar , 1992 .

[19]  Geoffrey E. Hinton,et al.  Keeping the neural networks simple by minimizing the description length of the weights , 1993, COLT '93.

[20]  Ali Mohammad-Djafari,et al.  Super-Resolution: A Short Review, A New Method Based on Hidden Markov Modeling of HR Image and Future Challenges , 2009, Comput. J..

[21]  Thomas Rodet,et al.  A Measure-Theoretic Variational Bayesian Algorithm for Large Dimensional Problems , 2014, SIAM J. Imaging Sci..

[22]  J. Giovannelli,et al.  Positive deconvolution for superimposed extended source and point sources , 2005, astro-ph/0507691.

[23]  Michael A. Saunders,et al.  Atomic Decomposition by Basis Pursuit , 1998, SIAM J. Sci. Comput..

[24]  F. Orieux,et al.  Super-resolution in map-making based on a physical instrument model and regularized inversion. Application to SPIRE/Herschel. , 2011, 1103.3698.

[25]  J. Högbom,et al.  APERTURE SYNTHESIS WITH A NON-REGULAR DISTRIBUTION OF INTERFEROMETER BASELINES. Commentary , 1974 .

[26]  Nikolas P. Galatsanos,et al.  Variational Bayesian Image Restoration Based on a Product of $t$-Distributions Image Prior , 2008, IEEE Transactions on Image Processing.

[27]  Barak A. Pearlmutter,et al.  Blind Source Separation by Sparse Decomposition in a Signal Dictionary , 2001, Neural Computation.

[28]  Moon Gi Kang,et al.  Super-resolution image reconstruction: a technical overview , 2003, IEEE Signal Process. Mag..

[29]  S. Ott,et al.  Herschel Space Observatory - An ESA facility for far-infrared and submillimetre astronomy , 2010, 1005.5331.

[30]  Jean-Luc Starck,et al.  Astronomical image and data analysis , 2002 .

[31]  S. J. Hewlett,et al.  Superresolution in confocal scanning microscopy. , 1991, Optics letters.