A novel approach to the selection of robust and invariant features for classification of hyperspectral images

Lorenzo Bruzzone; Claudio Persello

doi:10.1109/IGARSS.2008.4778794

A novel approach to the selection of robust and invariant features for classification of hyperspectral images

Lorenzo Bruzzone, Claudio Persello

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic › peer-review

1 Citation (Scopus)

Abstract

This paper presents a novel approach to feature selection for the classification of hyperspectral images. The proposed approach aims at selecting a subset of the original set of features that exhibits two main properties:( i) high capability to discriminate among the considered classes, (ii) high invariance (stationarity) in the spatial domain of the investigated scene. The feature selection is accomplished by defining a multi-objective criterion that considers two terms: (i) a term that assesses the class separability, (ii) a term that evaluates the spatial invariance of the selected features. The multi-objective problem is solved by an evolutionary algorithm that estimates the Pareto-optimal solutions. Experiments carried out on a hyperspectral image acquired by the Hyperion sensor confirmed the effectiveness of the proposed technique.

Original language	English
Title of host publication	Proceedings of IGARSS 2008
Subtitle of host publication	International Geoscience and Remote Sensing Symposium : Geoscience and remote sensing, the next generation, 6-11 July 2008 Boston, MA, USA.
Place of Publication	Piscataway, NJ, USA
Publisher	IEEE
Pages	I-66 - I-69
ISBN (Electronic)	978-1-4244-2808-3 (CD)
ISBN (Print)	978-1-4244-2807-6
DOIs	https://doi.org/10.1109/IGARSS.2008.4778794
Publication status	Published - 2008
Externally published	Yes
Event	2008 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2008 - Boston, United States Duration: 6 Jul 2008 → 11 Jul 2008

Conference

Conference	2008 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2008
Abbreviated title	IGARSS 2008
Country/Territory	United States
City	Boston
Period	6/07/08 → 11/07/08

Keywords

ADLIB-ART-323
n/a OA procedure

Access to Document

10.1109/IGARSS.2008.4778794

Cite this

Bruzzone, L., & Persello, C. (2008). A novel approach to the selection of robust and invariant features for classification of hyperspectral images. In Proceedings of IGARSS 2008: International Geoscience and Remote Sensing Symposium : Geoscience and remote sensing, the next generation, 6-11 July 2008 Boston, MA, USA. (pp. I-66 - I-69). IEEE. https://doi.org/10.1109/IGARSS.2008.4778794

Bruzzone, Lorenzo ; Persello, Claudio. / A novel approach to the selection of robust and invariant features for classification of hyperspectral images. Proceedings of IGARSS 2008: International Geoscience and Remote Sensing Symposium : Geoscience and remote sensing, the next generation, 6-11 July 2008 Boston, MA, USA.. Piscataway, NJ, USA : IEEE, 2008. pp. I-66 - I-69

@inbook{a9a07e6eff2a4f6aa6875e1f1249f61a,

title = "A novel approach to the selection of robust and invariant features for classification of hyperspectral images",

abstract = "This paper presents a novel approach to feature selection for the classification of hyperspectral images. The proposed approach aims at selecting a subset of the original set of features that exhibits two main properties:( i) high capability to discriminate among the considered classes, (ii) high invariance (stationarity) in the spatial domain of the investigated scene. The feature selection is accomplished by defining a multi-objective criterion that considers two terms: (i) a term that assesses the class separability, (ii) a term that evaluates the spatial invariance of the selected features. The multi-objective problem is solved by an evolutionary algorithm that estimates the Pareto-optimal solutions. Experiments carried out on a hyperspectral image acquired by the Hyperion sensor confirmed the effectiveness of the proposed technique.",

keywords = "ADLIB-ART-323, n/a OA procedure",

author = "Lorenzo Bruzzone and Claudio Persello",

year = "2008",

doi = "10.1109/IGARSS.2008.4778794",

language = "English",

isbn = "978-1-4244-2807-6",

pages = "I--66 -- I--69",

booktitle = "Proceedings of IGARSS 2008",

publisher = "IEEE",

address = "United States",

note = "2008 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2008, IGARSS 2008 ; Conference date: 06-07-2008 Through 11-07-2008",

}

Bruzzone, L & Persello, C 2008, A novel approach to the selection of robust and invariant features for classification of hyperspectral images. in Proceedings of IGARSS 2008: International Geoscience and Remote Sensing Symposium : Geoscience and remote sensing, the next generation, 6-11 July 2008 Boston, MA, USA.. IEEE, Piscataway, NJ, USA, pp. I-66 - I-69, 2008 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2008, Boston, Massachusetts, United States, 6/07/08. https://doi.org/10.1109/IGARSS.2008.4778794

A novel approach to the selection of robust and invariant features for classification of hyperspectral images. / Bruzzone, Lorenzo; Persello, Claudio.
Proceedings of IGARSS 2008: International Geoscience and Remote Sensing Symposium : Geoscience and remote sensing, the next generation, 6-11 July 2008 Boston, MA, USA.. Piscataway, NJ, USA: IEEE, 2008. p. I-66 - I-69.

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic › peer-review

TY - CHAP

T1 - A novel approach to the selection of robust and invariant features for classification of hyperspectral images

AU - Bruzzone, Lorenzo

AU - Persello, Claudio

PY - 2008

Y1 - 2008

N2 - This paper presents a novel approach to feature selection for the classification of hyperspectral images. The proposed approach aims at selecting a subset of the original set of features that exhibits two main properties:( i) high capability to discriminate among the considered classes, (ii) high invariance (stationarity) in the spatial domain of the investigated scene. The feature selection is accomplished by defining a multi-objective criterion that considers two terms: (i) a term that assesses the class separability, (ii) a term that evaluates the spatial invariance of the selected features. The multi-objective problem is solved by an evolutionary algorithm that estimates the Pareto-optimal solutions. Experiments carried out on a hyperspectral image acquired by the Hyperion sensor confirmed the effectiveness of the proposed technique.

AB - This paper presents a novel approach to feature selection for the classification of hyperspectral images. The proposed approach aims at selecting a subset of the original set of features that exhibits two main properties:( i) high capability to discriminate among the considered classes, (ii) high invariance (stationarity) in the spatial domain of the investigated scene. The feature selection is accomplished by defining a multi-objective criterion that considers two terms: (i) a term that assesses the class separability, (ii) a term that evaluates the spatial invariance of the selected features. The multi-objective problem is solved by an evolutionary algorithm that estimates the Pareto-optimal solutions. Experiments carried out on a hyperspectral image acquired by the Hyperion sensor confirmed the effectiveness of the proposed technique.

KW - ADLIB-ART-323

KW - n/a OA procedure

U2 - 10.1109/IGARSS.2008.4778794

DO - 10.1109/IGARSS.2008.4778794

M3 - Chapter

SN - 978-1-4244-2807-6

SP - I-66 - I-69

BT - Proceedings of IGARSS 2008

PB - IEEE

CY - Piscataway, NJ, USA

T2 - 2008 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2008

Y2 - 6 July 2008 through 11 July 2008

ER -

Bruzzone L, Persello C. A novel approach to the selection of robust and invariant features for classification of hyperspectral images. In Proceedings of IGARSS 2008: International Geoscience and Remote Sensing Symposium : Geoscience and remote sensing, the next generation, 6-11 July 2008 Boston, MA, USA.. Piscataway, NJ, USA: IEEE. 2008. p. I-66 - I-69 doi: 10.1109/IGARSS.2008.4778794

A novel approach to the selection of robust and invariant features for classification of hyperspectral images

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this