Robust unmixing algorithms for hyperspectral imagery

Abderrahim Halimi; Yoann Altmann; Gerald Stuart Buller; Stephen McLaughlin; William Oxford; Damien Clarke; Jonathan Piper

Robust unmixing algorithms for hyperspectral imagery

Abderrahim Halimi, Yoann Altmann, Gerald Stuart Buller, Stephen McLaughlin, William Oxford, Damien Clarke, Jonathan Piper

Research output: Contribution to conference › Paper › peer-review

116 Downloads (Pure)

Abstract

The linear mixture model (LMM) assumes a hyperspectral pixel
spectrum to be a linear combination of endmember spectra corrupted
by additive noise. This model is widely used for spectral
unmixing mainly because of its simplicity. However, the LMM
can be inappropriate in presence of nonlinear effects, endmember
variability or outliers. This paper presents a comparison between recent
robust hyperspectral unmixing algorithms. The mixture models
are first introduced followed by the description of their associated
unmixing algorithms. The algorithms are then analyzed when
considering a real image acquired over the region of Porton Down
in England. The results discuss the behavior of each algorithm to
unmix these data and compare their ability to detect the natural or
man-made outliers in the scene. The obtained results highlight the
potential of the studied mixture models to overcome the current
limitations of the LMM.

Original language	English
Publication status	Accepted/In press - 16 Jun 2016
Event	6th Sensor Signal Processing for Defence Conference 2016 - Edinburgh, United Kingdom Duration: 22 Sept 2016 → 23 Sept 2016 Conference number: 6th

Conference

Conference	6th Sensor Signal Processing for Defence Conference 2016
Abbreviated title	SSPD 2016
Country/Territory	United Kingdom
City	Edinburgh
Period	22/09/16 → 23/09/16

Access to Document

Download pdfAccepted author manuscript, 575 KB

Cite this

@conference{ff4059f51b4d46c893dc43f69d442b2a,

title = "Robust unmixing algorithms for hyperspectral imagery",

abstract = "The linear mixture model (LMM) assumes a hyperspectral pixelspectrum to be a linear combination of endmember spectra corruptedby additive noise. This model is widely used for spectralunmixing mainly because of its simplicity. However, the LMMcan be inappropriate in presence of nonlinear effects, endmembervariability or outliers. This paper presents a comparison between recentrobust hyperspectral unmixing algorithms. The mixture modelsare first introduced followed by the description of their associatedunmixing algorithms. The algorithms are then analyzed whenconsidering a real image acquired over the region of Porton Downin England. The results discuss the behavior of each algorithm tounmix these data and compare their ability to detect the natural orman-made outliers in the scene. The obtained results highlight thepotential of the studied mixture models to overcome the currentlimitations of the LMM.",

author = "Abderrahim Halimi and Yoann Altmann and Buller, {Gerald Stuart} and Stephen McLaughlin and William Oxford and Damien Clarke and Jonathan Piper",

year = "2016",

month = jun,

day = "16",

language = "English",

note = "6th Sensor Signal Processing for Defence Conference 2016, SSPD 2016 ; Conference date: 22-09-2016 Through 23-09-2016",

}

TY - CONF

T1 - Robust unmixing algorithms for hyperspectral imagery

AU - Halimi, Abderrahim

AU - Altmann, Yoann

AU - Buller, Gerald Stuart

AU - McLaughlin, Stephen

AU - Oxford, William

AU - Clarke, Damien

AU - Piper, Jonathan

N1 - Conference code: 6th

PY - 2016/6/16

Y1 - 2016/6/16

N2 - The linear mixture model (LMM) assumes a hyperspectral pixelspectrum to be a linear combination of endmember spectra corruptedby additive noise. This model is widely used for spectralunmixing mainly because of its simplicity. However, the LMMcan be inappropriate in presence of nonlinear effects, endmembervariability or outliers. This paper presents a comparison between recentrobust hyperspectral unmixing algorithms. The mixture modelsare first introduced followed by the description of their associatedunmixing algorithms. The algorithms are then analyzed whenconsidering a real image acquired over the region of Porton Downin England. The results discuss the behavior of each algorithm tounmix these data and compare their ability to detect the natural orman-made outliers in the scene. The obtained results highlight thepotential of the studied mixture models to overcome the currentlimitations of the LMM.

AB - The linear mixture model (LMM) assumes a hyperspectral pixelspectrum to be a linear combination of endmember spectra corruptedby additive noise. This model is widely used for spectralunmixing mainly because of its simplicity. However, the LMMcan be inappropriate in presence of nonlinear effects, endmembervariability or outliers. This paper presents a comparison between recentrobust hyperspectral unmixing algorithms. The mixture modelsare first introduced followed by the description of their associatedunmixing algorithms. The algorithms are then analyzed whenconsidering a real image acquired over the region of Porton Downin England. The results discuss the behavior of each algorithm tounmix these data and compare their ability to detect the natural orman-made outliers in the scene. The obtained results highlight thepotential of the studied mixture models to overcome the currentlimitations of the LMM.

M3 - Paper

T2 - 6th Sensor Signal Processing for Defence Conference 2016

Y2 - 22 September 2016 through 23 September 2016

ER -

Robust unmixing algorithms for hyperspectral imagery

Abstract

Conference

Access to Document

Fingerprint

Cite this