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 proceedingChapterAcademicpeer-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 languageEnglish
Title of host publicationProceedings of IGARSS 2008
Subtitle of host publicationInternational Geoscience and Remote Sensing Symposium : Geoscience and remote sensing, the next generation, 6-11 July 2008 Boston, MA, USA.
Place of PublicationPiscataway, NJ, USA
PublisherIEEE
PagesI-66 - I-69
ISBN (Electronic)978-1-4244-2808-3 (CD)
ISBN (Print)978-1-4244-2807-6
DOIs
Publication statusPublished - 2008
Externally publishedYes
Event2008 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2008 - Boston, United States
Duration: 6 Jul 200811 Jul 2008

Conference

Conference2008 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2008
Abbreviated titleIGARSS 2008
Country/TerritoryUnited States
CityBoston
Period6/07/0811/07/08

Keywords

  • ADLIB-ART-323
  • n/a OA procedure

Fingerprint

Dive into the research topics of 'A novel approach to the selection of robust and invariant features for classification of hyperspectral images'. Together they form a unique fingerprint.

Cite this