Introducing co-clustering for hyperspectral image analysis

Emma Izquierdo Verdiguier, Raul Zurita-Milla

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

2 Citations (Scopus)

Abstract

This work introduces the use of co-clustering for hyperspectral image analysis. Co-clustering is able to simultaneously group samples (rows) and spectral bands (columns). This results in blocks, which do not only share spectral information (classical one way clustering) but also share sample information. Here, we propose using a co-clustering algorithm based on Information Theory - the optimal co-clustering is obtaining minimizing the loss of information between the original and the co-clustered images. A hyperspectral image (160000 samples and 40 bands) is used to illustrate this study. This image was clustered into 150 groups (50 groups of samples and 3 spectral groups). After that, blocks of the spectral groups was independently classified to assess the effectiveness of the co-clustering approach for hyperspectral band selection applications. Furthermore, the results were also compared with state-of-art methods based on morphological profiles, and the covariance matrix of the original hyperspectral image. Good results were achieved, showing the effectiveness of the Co-clustering approach for hyperspectral images in spatial-spectral classification and band selection applications.
Original languageEnglish
Title of host publication2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS)
Place of PublicationPiscataway, NJ
PublisherIEEE
Pages465-468
Number of pages4
ISBN (Electronic)978-1-4799-7929-5
ISBN (Print)978-1-4799-7928-8
DOIs
Publication statusPublished - 2015
EventIEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015: Remote Sensing: Understanding the Earth for a Safer World - Milan, Italy
Duration: 26 Jul 201531 Jul 2015
http://www.igarss2015.org/

Publication series

NameIEEE International Geoscience and Remote Sensing Symposium
PublisherIEEE
ISSN (Print)2153-6996
ISSN (Electronic)2153-7003

Conference

ConferenceIEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015
Abbreviated titleIGARSS 2015
CountryItaly
CityMilan
Period26/07/1531/07/15
Internet address

Fingerprint

image analysis
matrix

Keywords

  • METIS-316405

Cite this

Izquierdo Verdiguier, E., & Zurita-Milla, R. (2015). Introducing co-clustering for hyperspectral image analysis. In 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) (pp. 465-468). (IEEE International Geoscience and Remote Sensing Symposium). Piscataway, NJ: IEEE. https://doi.org/10.1109/IGARSS.2015.7325801
Izquierdo Verdiguier, Emma ; Zurita-Milla, Raul. / Introducing co-clustering for hyperspectral image analysis. 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). Piscataway, NJ : IEEE, 2015. pp. 465-468 (IEEE International Geoscience and Remote Sensing Symposium).
@inbook{1cb242b83f7e42bb881c90d6f2553171,
title = "Introducing co-clustering for hyperspectral image analysis",
abstract = "This work introduces the use of co-clustering for hyperspectral image analysis. Co-clustering is able to simultaneously group samples (rows) and spectral bands (columns). This results in blocks, which do not only share spectral information (classical one way clustering) but also share sample information. Here, we propose using a co-clustering algorithm based on Information Theory - the optimal co-clustering is obtaining minimizing the loss of information between the original and the co-clustered images. A hyperspectral image (160000 samples and 40 bands) is used to illustrate this study. This image was clustered into 150 groups (50 groups of samples and 3 spectral groups). After that, blocks of the spectral groups was independently classified to assess the effectiveness of the co-clustering approach for hyperspectral band selection applications. Furthermore, the results were also compared with state-of-art methods based on morphological profiles, and the covariance matrix of the original hyperspectral image. Good results were achieved, showing the effectiveness of the Co-clustering approach for hyperspectral images in spatial-spectral classification and band selection applications.",
keywords = "METIS-316405",
author = "{Izquierdo Verdiguier}, Emma and Raul Zurita-Milla",
year = "2015",
doi = "10.1109/IGARSS.2015.7325801",
language = "English",
isbn = "978-1-4799-7928-8",
series = "IEEE International Geoscience and Remote Sensing Symposium",
publisher = "IEEE",
pages = "465--468",
booktitle = "2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS)",
address = "United States",

}

Izquierdo Verdiguier, E & Zurita-Milla, R 2015, Introducing co-clustering for hyperspectral image analysis. in 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). IEEE International Geoscience and Remote Sensing Symposium, IEEE, Piscataway, NJ, pp. 465-468, IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2015, Milan, Italy, 26/07/15. https://doi.org/10.1109/IGARSS.2015.7325801

Introducing co-clustering for hyperspectral image analysis. / Izquierdo Verdiguier, Emma ; Zurita-Milla, Raul.

2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). Piscataway, NJ : IEEE, 2015. p. 465-468 (IEEE International Geoscience and Remote Sensing Symposium).

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

TY - CHAP

T1 - Introducing co-clustering for hyperspectral image analysis

AU - Izquierdo Verdiguier, Emma

AU - Zurita-Milla, Raul

PY - 2015

Y1 - 2015

N2 - This work introduces the use of co-clustering for hyperspectral image analysis. Co-clustering is able to simultaneously group samples (rows) and spectral bands (columns). This results in blocks, which do not only share spectral information (classical one way clustering) but also share sample information. Here, we propose using a co-clustering algorithm based on Information Theory - the optimal co-clustering is obtaining minimizing the loss of information between the original and the co-clustered images. A hyperspectral image (160000 samples and 40 bands) is used to illustrate this study. This image was clustered into 150 groups (50 groups of samples and 3 spectral groups). After that, blocks of the spectral groups was independently classified to assess the effectiveness of the co-clustering approach for hyperspectral band selection applications. Furthermore, the results were also compared with state-of-art methods based on morphological profiles, and the covariance matrix of the original hyperspectral image. Good results were achieved, showing the effectiveness of the Co-clustering approach for hyperspectral images in spatial-spectral classification and band selection applications.

AB - This work introduces the use of co-clustering for hyperspectral image analysis. Co-clustering is able to simultaneously group samples (rows) and spectral bands (columns). This results in blocks, which do not only share spectral information (classical one way clustering) but also share sample information. Here, we propose using a co-clustering algorithm based on Information Theory - the optimal co-clustering is obtaining minimizing the loss of information between the original and the co-clustered images. A hyperspectral image (160000 samples and 40 bands) is used to illustrate this study. This image was clustered into 150 groups (50 groups of samples and 3 spectral groups). After that, blocks of the spectral groups was independently classified to assess the effectiveness of the co-clustering approach for hyperspectral band selection applications. Furthermore, the results were also compared with state-of-art methods based on morphological profiles, and the covariance matrix of the original hyperspectral image. Good results were achieved, showing the effectiveness of the Co-clustering approach for hyperspectral images in spatial-spectral classification and band selection applications.

KW - METIS-316405

UR - https://ezproxy2.utwente.nl/login?url=https://webapps.itc.utwente.nl/library/2015/chap/zuritamilla_int.pdf

U2 - 10.1109/IGARSS.2015.7325801

DO - 10.1109/IGARSS.2015.7325801

M3 - Chapter

SN - 978-1-4799-7928-8

T3 - IEEE International Geoscience and Remote Sensing Symposium

SP - 465

EP - 468

BT - 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

PB - IEEE

CY - Piscataway, NJ

ER -

Izquierdo Verdiguier E, Zurita-Milla R. Introducing co-clustering for hyperspectral image analysis. In 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS). Piscataway, NJ: IEEE. 2015. p. 465-468. (IEEE International Geoscience and Remote Sensing Symposium). https://doi.org/10.1109/IGARSS.2015.7325801