Centralized collaborative sparse unmixing for hyperspectral images

Rui Wang, Heng-Chao Li, Wenzhi Liao, Xin Huang, Wilfried Philips

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Spectral unmixing is very important in hyperspectral image analysis and processing, which aims at identifying the constituent spectra (i.e., endmembers) and estimating their fractional abundances from the mixed pixels. In recent years, sparse unmixing has received considerable interest. However, the acquired hyperspectral images are generally degraded by the noise, making sparse unmixing not faithful enough. To address this issue, this paper proposes a novel framework to couple sparse hyperspectral unmixing and abundance estimation error reduction together. Specifically, with the definition of abundance estimation error, a centralized constraint is incorporated into the collaborative sparse unmixing framework by exploiting the nonlocal redundancy of abundance map. This way we suppress the abundance estimation error, and improve the unmixing accuracy. Meanwhile, the alternating direction method of multipliers is introduced to solve the underlying constrained model. Experimental results on both synthetic and real hyperspectral data demonstrate the effectiveness of our proposed algorithm.
LanguageEnglish
Pages1949-1962
Number of pages14
JournalIEEE Journal of Selected Topics in Earth Observation and Remote Sensing
Volume10
Issue number5
DOIs
Publication statusPublished - 6 Feb 2017

Fingerprint

abundance estimation
Error analysis
image processing
image analysis
Image analysis
Redundancy
pixel
Image processing
Pixels
multipliers
redundancy
estimating
pixels

Keywords

  • abundance estimation error
  • hyperspectral images
  • nonlocal means (NLM)
  • spectral unmixing
  • estimation error
  • collaboration
  • algorithm design and analysis
  • optimization
  • remote sensing

Cite this

Wang, Rui ; Li, Heng-Chao ; Liao, Wenzhi ; Huang, Xin ; Philips, Wilfried. / Centralized collaborative sparse unmixing for hyperspectral images. In: IEEE Journal of Selected Topics in Earth Observation and Remote Sensing. 2017 ; Vol. 10, No. 5. pp. 1949-1962.
@article{c877353854404d71be5a0a9f9c7f2c18,
title = "Centralized collaborative sparse unmixing for hyperspectral images",
abstract = "Spectral unmixing is very important in hyperspectral image analysis and processing, which aims at identifying the constituent spectra (i.e., endmembers) and estimating their fractional abundances from the mixed pixels. In recent years, sparse unmixing has received considerable interest. However, the acquired hyperspectral images are generally degraded by the noise, making sparse unmixing not faithful enough. To address this issue, this paper proposes a novel framework to couple sparse hyperspectral unmixing and abundance estimation error reduction together. Specifically, with the definition of abundance estimation error, a centralized constraint is incorporated into the collaborative sparse unmixing framework by exploiting the nonlocal redundancy of abundance map. This way we suppress the abundance estimation error, and improve the unmixing accuracy. Meanwhile, the alternating direction method of multipliers is introduced to solve the underlying constrained model. Experimental results on both synthetic and real hyperspectral data demonstrate the effectiveness of our proposed algorithm.",
keywords = "abundance estimation error, hyperspectral images, nonlocal means (NLM), spectral unmixing, estimation error, collaboration, algorithm design and analysis, optimization, remote sensing",
author = "Rui Wang and Heng-Chao Li and Wenzhi Liao and Xin Huang and Wilfried Philips",
year = "2017",
month = "2",
day = "6",
doi = "10.1109/JSTARS.2017.2651063",
language = "English",
volume = "10",
pages = "1949--1962",
journal = "IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing",
issn = "1939-1404",
publisher = "IEEE",
number = "5",

}

Centralized collaborative sparse unmixing for hyperspectral images. / Wang, Rui; Li, Heng-Chao; Liao, Wenzhi; Huang, Xin; Philips, Wilfried.

In: IEEE Journal of Selected Topics in Earth Observation and Remote Sensing, Vol. 10, No. 5, 06.02.2017, p. 1949-1962.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Centralized collaborative sparse unmixing for hyperspectral images

AU - Wang, Rui

AU - Li, Heng-Chao

AU - Liao, Wenzhi

AU - Huang, Xin

AU - Philips, Wilfried

PY - 2017/2/6

Y1 - 2017/2/6

N2 - Spectral unmixing is very important in hyperspectral image analysis and processing, which aims at identifying the constituent spectra (i.e., endmembers) and estimating their fractional abundances from the mixed pixels. In recent years, sparse unmixing has received considerable interest. However, the acquired hyperspectral images are generally degraded by the noise, making sparse unmixing not faithful enough. To address this issue, this paper proposes a novel framework to couple sparse hyperspectral unmixing and abundance estimation error reduction together. Specifically, with the definition of abundance estimation error, a centralized constraint is incorporated into the collaborative sparse unmixing framework by exploiting the nonlocal redundancy of abundance map. This way we suppress the abundance estimation error, and improve the unmixing accuracy. Meanwhile, the alternating direction method of multipliers is introduced to solve the underlying constrained model. Experimental results on both synthetic and real hyperspectral data demonstrate the effectiveness of our proposed algorithm.

AB - Spectral unmixing is very important in hyperspectral image analysis and processing, which aims at identifying the constituent spectra (i.e., endmembers) and estimating their fractional abundances from the mixed pixels. In recent years, sparse unmixing has received considerable interest. However, the acquired hyperspectral images are generally degraded by the noise, making sparse unmixing not faithful enough. To address this issue, this paper proposes a novel framework to couple sparse hyperspectral unmixing and abundance estimation error reduction together. Specifically, with the definition of abundance estimation error, a centralized constraint is incorporated into the collaborative sparse unmixing framework by exploiting the nonlocal redundancy of abundance map. This way we suppress the abundance estimation error, and improve the unmixing accuracy. Meanwhile, the alternating direction method of multipliers is introduced to solve the underlying constrained model. Experimental results on both synthetic and real hyperspectral data demonstrate the effectiveness of our proposed algorithm.

KW - abundance estimation error

KW - hyperspectral images

KW - nonlocal means (NLM)

KW - spectral unmixing

KW - estimation error

KW - collaboration

KW - algorithm design and analysis

KW - optimization

KW - remote sensing

UR - http://hdl.handle.net/1854/LU-8507924

U2 - 10.1109/JSTARS.2017.2651063

DO - 10.1109/JSTARS.2017.2651063

M3 - Article

VL - 10

SP - 1949

EP - 1962

JO - IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing

T2 - IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing

JF - IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing

SN - 1939-1404

IS - 5

ER -