Total variation and ℓq based hyperspectral unmixing for feature extraction and classification

Jakob Sigurdsson; Magnus O. Ulfarsson; Johannes R. Sveinsson

doi:10.1109/IGARSS.2015.7325794

Total variation and ℓ_q based hyperspectral unmixing for feature extraction and classification

Jakob Sigurdsson
, Magnus O. Ulfarsson
, Johannes R. Sveinsson

Rafmagns- og tölvuverkfræðideild

Háskóli Íslands

Rannsóknarafurð: Kafli í bók/skýrslu/ráðstefnuriti › Ráðstefnuframlag › ritrýni

Útdráttur

Blind hyperspectral unmixing jointly estimates both the endmembers and the abundances of hyperspectral images. The endmembers represent the spectral signatures of material found in the image and the abundances specify the amount of each material seen in each pixel in the image. In this paper, a blind hyperspectral unmixing method for feature extraction and classification using total variation (TV) and ℓ_q sparse regularization is proposed. The abundances found are used as features for classification. The classification results are compared to results obtained using Principal Component analysis (PCA) and also to results obtained using hyperspectral unmixing using only TV and sparsity, respectively.

Upprunalegt tungumál	Enska
Titill gistiútgáfu	2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings
Útgefandi	Institute of Electrical and Electronics Engineers Inc.
Síður	437-440
Síðufjöldi	4
ISBN-númer (rafrænt)	9781479979295
ISBN-númer (prentað)	9781479979295
DOI	https://doi.org/10.1109/IGARSS.2015.7325794
Útgáfustaða	Útgefið - 10 nóv. 2015
Viðburður	IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Milan, Ítalía Tímalengd: 26 júl. 2015 → 31 júl. 2015

Ritröð

Nafn	International Geoscience and Remote Sensing Symposium (IGARSS)
Bindi	2015-November

Ráðstefna

Ráðstefna	IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015
Land/Yfirráðasvæði	Ítalía
Borg/bær	Milan
Tímabil	26/07/15 → 31/07/15

Athugasemd

Aðgangur að skjali

10.1109/IGARSS.2015.7325794

Vitna í þetta

Sigurdsson, J., Ulfarsson, M. O., & Sveinsson, J. R. (2015). Total variation and ℓ_q based hyperspectral unmixing for feature extraction and classification. Í 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings (bls. 437-440). Grein 7325794 (International Geoscience and Remote Sensing Symposium (IGARSS); Bindi 2015-November). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IGARSS.2015.7325794

Sigurdsson, Jakob ; Ulfarsson, Magnus O. ; Sveinsson, Johannes R. / Total variation and ℓ_q based hyperspectral unmixing for feature extraction and classification. 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. bls. 437-440 (International Geoscience and Remote Sensing Symposium (IGARSS)).

@inproceedings{397247bd17534bf184bd4775a9f43890,

title = "Total variation and ℓq based hyperspectral unmixing for feature extraction and classification",

abstract = "Blind hyperspectral unmixing jointly estimates both the endmembers and the abundances of hyperspectral images. The endmembers represent the spectral signatures of material found in the image and the abundances specify the amount of each material seen in each pixel in the image. In this paper, a blind hyperspectral unmixing method for feature extraction and classification using total variation (TV) and ℓq sparse regularization is proposed. The abundances found are used as features for classification. The classification results are compared to results obtained using Principal Component analysis (PCA) and also to results obtained using hyperspectral unmixing using only TV and sparsity, respectively.",

keywords = "Hyperspectral unmixing, blind signal separation, classification, dyadic cyclic descent, feature extraction, linear unmixing, total variation",

author = "Jakob Sigurdsson and Ulfarsson, \{Magnus O.\} and Sveinsson, \{Johannes R.\}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 ; Conference date: 26-07-2015 Through 31-07-2015",

year = "2015",

month = nov,

day = "10",

doi = "10.1109/IGARSS.2015.7325794",

language = "English",

isbn = "9781479979295",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "437--440",

booktitle = "2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings",

address = "United States",

}

Sigurdsson, J , Ulfarsson, MO & Sveinsson, JR 2015, Total variation and ℓ_q based hyperspectral unmixing for feature extraction and classification. í 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings., 7325794, International Geoscience and Remote Sensing Symposium (IGARSS), bind. 2015-November, Institute of Electrical and Electronics Engineers Inc., bls. 437-440, IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015, Milan, Ítalía, 26/07/15. https://doi.org/10.1109/IGARSS.2015.7325794

Total variation and ℓ_q based hyperspectral unmixing for feature extraction and classification. / Sigurdsson, Jakob ; Ulfarsson, Magnus O.; Sveinsson, Johannes R.
2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. s. 437-440 7325794 (International Geoscience and Remote Sensing Symposium (IGARSS); Bindi 2015-November).

Rannsóknarafurð: Kafli í bók/skýrslu/ráðstefnuriti › Ráðstefnuframlag › ritrýni

TY - GEN

T1 - Total variation and ℓq based hyperspectral unmixing for feature extraction and classification

AU - Sigurdsson, Jakob

AU - Ulfarsson, Magnus O.

AU - Sveinsson, Johannes R.

PY - 2015/11/10

Y1 - 2015/11/10

N2 - Blind hyperspectral unmixing jointly estimates both the endmembers and the abundances of hyperspectral images. The endmembers represent the spectral signatures of material found in the image and the abundances specify the amount of each material seen in each pixel in the image. In this paper, a blind hyperspectral unmixing method for feature extraction and classification using total variation (TV) and ℓq sparse regularization is proposed. The abundances found are used as features for classification. The classification results are compared to results obtained using Principal Component analysis (PCA) and also to results obtained using hyperspectral unmixing using only TV and sparsity, respectively.

AB - Blind hyperspectral unmixing jointly estimates both the endmembers and the abundances of hyperspectral images. The endmembers represent the spectral signatures of material found in the image and the abundances specify the amount of each material seen in each pixel in the image. In this paper, a blind hyperspectral unmixing method for feature extraction and classification using total variation (TV) and ℓq sparse regularization is proposed. The abundances found are used as features for classification. The classification results are compared to results obtained using Principal Component analysis (PCA) and also to results obtained using hyperspectral unmixing using only TV and sparsity, respectively.

KW - Hyperspectral unmixing

KW - blind signal separation

KW - classification

KW - dyadic cyclic descent

KW - feature extraction

KW - linear unmixing

KW - total variation

UR - https://www.scopus.com/pages/publications/84962483131

U2 - 10.1109/IGARSS.2015.7325794

DO - 10.1109/IGARSS.2015.7325794

M3 - Conference contribution

SN - 9781479979295

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 437

EP - 440

BT - 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015

Y2 - 26 July 2015 through 31 July 2015

ER -

Sigurdsson J , Ulfarsson MO, Sveinsson JR. Total variation and ℓ_q based hyperspectral unmixing for feature extraction and classification. Í 2015 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2015. s. 437-440. 7325794. (International Geoscience and Remote Sensing Symposium (IGARSS)). doi: 10.1109/IGARSS.2015.7325794