OATAO - Open Archive Toulouse Archive Ouverte Open Access Week

Nonparametric Detection of Nonlinearly Mixed Pixels and Endmember Estimation in Hyperspectral Images

Imbiriba, Tales and Bermudez, José Carlos Moreira and Richard, Cédric and Tourneret, Jean-Yves Nonparametric Detection of Nonlinearly Mixed Pixels and Endmember Estimation in Hyperspectral Images. (2015) IEEE Transactions on Image Processing, vol. 25 (n° 3). pp. 1136-1151. ISSN 1057-7149

[img]
Preview
(Document in English)

PDF (Author's version) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
4MB

Official URL: http://dx.doi.org/10.1109/TIP.2015.2509258

Abstract

Mixing phenomena in hyperspectral images depend on a variety of factors, such as the resolution of observation devices, the properties of materials, and how these materials interact with incident light in the scene. Different parametric and nonparametric models have been considered to address hyperspectral unmixing problems. The simplest one is the linear mixing model. Nevertheless, it has been recognized that the mixing phenomena can also be nonlinear. The corresponding nonlinear analysis techniques are necessarily more challenging and complex than those employed for linear unmixing. Within this context, it makes sense to detect the nonlinearly mixed pixels in an image prior to its analysis, and then employ the simplest possible unmixing technique to analyze each pixel. In this paper, we propose a technique for detecting nonlinearly mixed pixels. The detection approach is based on the comparison of the reconstruction errors using both a Gaussian process regression model and a linear regression model. The two errors are combined into a detection statistics for which a probability density function can be reasonably approximated. We also propose an iterative endmember extraction algorithm to be employed in combination with the detection algorithm. The proposed detect-then-unmix strategy, which consists of extracting endmembers, detecting nonlinearly mixed pixels and unmixing, is tested with synthetic and real images.

Item Type:Article
Additional Information:Thanks to IEEE editor. The definitive version is available at http://ieeexplore.ieee.org/Xplore/home.jsp.com The original PDF of the article can be found at http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7358107
HAL Id:hal-01315353
Audience (journal):International peer-reviewed journal
Uncontrolled Keywords:
Institution:French research institutions > Centre National de la Recherche Scientifique - CNRS (FRANCE)
Université de Toulouse > Institut National Polytechnique de Toulouse - INPT (FRANCE)
Other partners > Université Nice Sophia Antipolis (FRANCE)
Other partners > Observatoire de la Côte d'Azur (FRANCE)
Other partners > Universidade Federal de Santa Catarina - UFSC (BRAZIL)
Université de Toulouse > Université Toulouse III - Paul Sabatier - UPS (FRANCE)
Université de Toulouse > Université Toulouse - Jean Jaurès - UT2J (FRANCE)
Université de Toulouse > Université Toulouse 1 Capitole - UT1 (FRANCE)
Laboratory name:
Statistics:download
Deposited By: Jean-yves TOURNERET
Deposited On:11 May 2016 10:17

Repository Staff Only: item control page