Improved Local Spectral Unmixing of hyperspectral data using an algorithmic regularization path for collaborative sparse regression

Local Spectral Unmixing (LSU) methods perform the unmixing of hyperspectral data locally in regions of the image. The endmembers and their abundances in each pixel are extracted region-wise, instead of globally to mitigate spectral variability effects, which are less severe locally. However, it requires the local estimation of the number of endmembers to use. Algorithms for intrinsic dimensionality (ID) estimation tend to overestimate the local ID, especially in small regions. The ID only provides an upper bound of the application and scale dependent number of endmembers, which leads to extract irrelevant signatures as local endmembers, associated with meaningless local abundances. We propose a method to select in each region the best subset of the locally extracted endmembers. Collaborative sparsity is used to detect spurious endmembers in each region and only keep the most influent ones. We compute an algorithmic regularization path for this problem, giving access to the sequence of successive active sets of endmembers when the regularization parameter is increased. Finally, we select the optimal set in the sense of the Bayesian Information Criterion (BIC), favoring models with a high likelihood, while penalizing those with too many endmembers. Results on real data show the interest of the proposed approach.

[1]  Antonio J. Plaza,et al.  Hyperspectral Image Segmentation Using a New Spectral Unmixing-Based Binary Partition Tree Representation , 2014, IEEE Transactions on Image Processing.

[2]  Silvia Valero,et al.  Hyperspectral Image Representation and Processing With Binary Partition Trees , 2013, IEEE Transactions on Image Processing.

[3]  G. Schwarz Estimating the Dimension of a Model , 1978 .

[4]  David Mary,et al.  Blind and Fully Constrained Unmixing of Hyperspectral Images , 2014, IEEE Transactions on Image Processing.

[5]  Laurent Condat Fast projection onto the simplex and the l1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\pmb {l}_\mathbf {1}$$\end{ , 2015, Mathematical Programming.

[6]  Antonio J. Plaza,et al.  Collaborative Sparse Regression for Hyperspectral Unmixing , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[7]  D. B. Preston Spectral Analysis and Time Series , 1983 .

[8]  M. Kowalski Sparse regression using mixed norms , 2009 .

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

[10]  Jean-Yves Tourneret,et al.  Hyperspectral Unmixing With Spectral Variability Using a Perturbed Linear Mixing Model , 2015, IEEE Transactions on Signal Processing.

[11]  Paul D. Gader,et al.  A Review of Nonlinear Hyperspectral Unmixing Methods , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[12]  Amandine Robin,et al.  Determining the Intrinsic Dimension of a Hyperspectral Image Using Random Matrix Theory , 2013, IEEE Transactions on Image Processing.

[13]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Manuel Graña,et al.  An endmember-based distance for content based hyperspectral image retrieval , 2012, Pattern Recognit..

[15]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[16]  Jocelyn Chanussot,et al.  Blind Hyperspectral Unmixing Using an Extended Linear Mixing Model to Address Spectral Variability , 2016, IEEE Transactions on Image Processing.

[17]  Asad Mahmood,et al.  Estimation of the Intrinsic Dimension of Hyperspectral Images: Comparison of Current Methods , 2015, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[18]  Jocelyn Chanussot,et al.  Hyperspectral Local Intrinsic Dimensionality , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[19]  Jocelyn Chanussot,et al.  Dynamical Spectral Unmixing of Multitemporal Hyperspectral Images , 2015, IEEE Transactions on Image Processing.

[20]  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.

[21]  Maria C. Torres-Madronero,et al.  Unmixing Analysis of a Time Series of Hyperion Images Over the Guánica Dry Forest in Puerto Rico , 2013, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[22]  Genevera I. Allen,et al.  ADMM Algorithmic Regularization Paths for Sparse Statistical Machine Learning , 2015, 1504.06637.

[23]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[24]  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..

[25]  David W. Messinger,et al.  Spatially Adaptive Hyperspectral Unmixing , 2011, IEEE Transactions on Geoscience and Remote Sensing.

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

[27]  Paul D. Gader,et al.  Signal and Image Processing in Hyperspectral Remote Sensing [From the Guest Editors] , 2014 .