Unsupervised Post-Nonlinear Unmixing of Hyperspectral Images Using a Hamiltonian Monte Carlo Algorithm

This paper presents a nonlinear mixing model for hyperspectral image unmixing. The proposed model assumes that the pixel reflectances are post-nonlinear functions of unknown pure spectral components contaminated by an additive white Gaussian noise. These nonlinear functions are approximated using second-order polynomials leading to a polynomial post-nonlinear mixing model. A Bayesian algorithm is proposed to estimate the parameters involved in the model yielding an unsupervised nonlinear unmixing algorithm. Due to the large number of parameters to be estimated, an efficient Hamiltonian Monte Carlo algorithm is investigated. The classical leapfrog steps of this algorithm are modified to handle the parameter constraints. The performance of the unmixing strategy, including convergence and parameter tuning, is first evaluated on synthetic data. Simulations conducted with real data finally show the accuracy of the proposed unmixing strategy for the analysis of hyperspectral images.

[1]  Carsten Hartmann,et al.  A constrained hybrid Monte‐Carlo algorithm and the problem of calculating the free energy in several variables , 2005 .

[2]  Pol Coppin,et al.  Endmember variability in Spectral Mixture Analysis: A review , 2011 .

[3]  Mario Winter,et al.  N-FINDR: an algorithm for fast autonomous spectral end-member determination in hyperspectral data , 1999, Optics & Photonics.

[4]  Mikkel N. Schmidt.  Function factorization using warped Gaussian processes , 2009, ICML '09.

[5]  Chein-I Chang,et al.  Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imagery , 2001, IEEE Trans. Geosci. Remote. Sens..

[6]  Jean-Yves Tourneret,et al.  Nonlinear Spectral Unmixing of Hyperspectral Images Using Gaussian Processes , 2012, IEEE Transactions on Signal Processing.

[7]  José M. Bioucas-Dias,et al.  Hyperspectral Subspace Identification , 2008, IEEE Transactions on Geoscience and Remote Sensing.

[8]  Chong-Yung Chi,et al.  Hyperspectral Data Geometry-Based Estimation of Number of Endmembers Using p-Norm-Based Pure Pixel Identification Algorithm , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[9]  B. Hapke.  Bidirectional reflectance spectroscopy , 1984 .

[10]  Amit Banerjee,et al.  Automated endmember determination and adaptive spectral mixture analysis using kernel methods , 2013, Optics & Photonics - Optical Engineering + Applications.

[11]  John R. Miller,et al.  Comparative study between a new nonlinear model and common linear model for analysing laboratory simulated‐forest hyperspectral data , 2009 .

[12]  Geoffrey E. Hinton,et al.  Bayesian Learning for Neural Networks , 1995 .

[13]  Paul D. Gader,et al.  Bootstrapping for Piece-Wise Convex Endmember Distribution Detection , 2012, 2012 4th Workshop on Hyperspectral Image and Signal Processing (WHISPERS).

[14]  Andrew Gelman,et al.  Handbook of Markov Chain Monte Carlo , 2011 .

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

[16]  Jean-Yves Tourneret,et al.  Nonlinear unmixing of hyperspectral images using a generalized bilinear model , 2011, 2011 IEEE Statistical Signal Processing Workshop (SSP).

[17]  Chein-I Chang,et al.  A quantitative and comparative analysis of linear and nonlinear spectral mixture models using radial basis function neural networks , 2001, IEEE Trans. Geosci. Remote. Sens..

[18]  K. C. Ho,et al.  Endmember Variability in Hyperspectral Analysis: Addressing Spectral Variability During Spectral Unmixing , 2014, IEEE Signal Processing Magazine.

[19]  Jie Chen,et al.  Estimating abundance fractions of materials in hyperspectral images by fitting a post-nonlinear mixing model , 2013, 2013 5th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS).

[20]  Shengli Xie,et al.  Blind Spectral Unmixing Based on Sparse Nonnegative Matrix Factorization , 2011, IEEE Transactions on Image Processing.

[21]  Mathieu Fauvel,et al.  Mapping ash tree colonization in an agricultural mountain landscape: Investigating the potential of hyperspectral imagery , 2011, 2011 IEEE International Geoscience and Remote Sensing Symposium.

[22]  Jean-Yves Tourneret,et al.  Supervised Nonlinear Spectral Unmixing Using a Postnonlinear Mixing Model for Hyperspectral Imagery , 2012, IEEE Transactions on Image Processing.

[23]  B. Hapke.  Bidirectional reflectance spectroscopy: 1. Theory , 1981 .

[24]  J. Karhunen,et al.  Advances in Nonlinear Blind Source Separation , 2003 .

[25]  Maurice D. Craig,et al.  Minimum-volume transforms for remotely sensed data , 1994, IEEE Trans. Geosci. Remote. Sens..

[26]  Raquel Urtasun,et al.  A Family of MCMC Methods on Implicitly Defined Manifolds , 2012, AISTATS.

[27]  Hairong Qi,et al.  A Maximum Entropy Approach to Unsupervised Mixed-Pixel Decomposition , 2007, IEEE Transactions on Image Processing.

[28]  M. Betancourt.  Cruising The Simplex: Hamiltonian Monte Carlo and the Dirichlet Distribution , 2010, 1010.3436.

[29]  Antonio J. Plaza,et al.  Hyperspectral Unmixing Overview: Geometrical, Statistical, and Sparse Regression-Based Approaches , 2012, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[30]  Jie Chen,et al.  Nonlinear Unmixing of Hyperspectral Data Based on a Linear-Mixture/Nonlinear-Fluctuation Model , 2013, IEEE Transactions on Signal Processing.

[31]  S. Duane,et al.  Hybrid Monte Carlo , 1987 .

[32]  Jean-Yves Tourneret,et al.  Nonlinear unmixing of hyperspectral images using radial basis functions and orthogonal least squares , 2011, 2011 IEEE International Geoscience and Remote Sensing Symposium.

[33]  Naoto Yokoya,et al.  Nonlinear Unmixing of Hyperspectral Data Using Semi-Nonnegative Matrix Factorization , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[34]  Trac D. Tran,et al.  Low rank representation for bilinear abundance estimation problem , 2013, 2013 5th Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS).

[35]  Alfonso Fernández-Manso,et al.  Spectral unmixing , 2012 .

[36]  Alfred O. Hero,et al.  Joint Bayesian Endmember Extraction and Linear Unmixing for Hyperspectral Imagery , 2009, IEEE Transactions on Signal Processing.

[37]  José M. Bioucas-Dias,et al.  Vertex component analysis: a fast algorithm to unmix hyperspectral data , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[38]  Hsuan Ren,et al.  Extended linear hyperspectral mixing models , 2004, SPIE Optics East.

[39]  Rob Heylen,et al.  Non-Linear Spectral Unmixing by Geodesic Simplex Volume Maximization , 2011, IEEE Journal of Selected Topics in Signal Processing.

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

[41]  Hoon Kim,et al.  Monte Carlo Statistical Methods , 2000, Technometrics.

[42]  Jean-Yves Tourneret,et al.  Unmixing hyperspectral images using the generalized bilinear model , 2011, 2011 IEEE International Geoscience and Remote Sensing Symposium.

[43]  Christian P. Robert,et al.  Convergence Control of MCMC Algorithms , 1998 .

[44]  Christian Jutten,et al.  BLIND SEPARATING CONVOLUTIVE POST NON-LINEAR MIXTURES , 2001 .

[45]  Alfred O. Hero,et al.  Nonlinear Unmixing of Hyperspectral Images: Models and Algorithms , 2013, IEEE Signal Processing Magazine.

[46]  Jean-Yves Tourneret,et al.  Bayesian Estimation of Linear Mixtures Using the Normal Compositional Model. Application to Hyperspectral Imagery , 2010, IEEE Transactions on Image Processing.

[47]  Chein-I Chang,et al.  Pixel purity index-based algorithms for endmember extraction from hyperspectral imagery , 2006 .

[48]  José M. Bioucas-Dias,et al.  Nonlinear mixture model for hyperspectral unmixing , 2009, Remote Sensing.

[49]  Joseph N. Wilson,et al.  Using physics-based macroscopic and microscopic mixture models for hyperspectral pixel unmixing , 2012, Defense + Commercial Sensing.