Feature selection for classification of hyperspectral data by SVM

SVM are attractive for the classification of remotely sensed data with some claims that the method is insensitive to the dimensionality of the data and so not requiring a dimensionality reduction analysis in pre-processing. Here, a series of classification analyses with two hyperspectral sensor data...

Full description

Bibliographic Details
Main Authors: Pal, Mahesh, Foody, Giles M.
Format: Article
Published: Institute of Electrical and Electronics Engineers 2010
Online Access:https://eprints.nottingham.ac.uk/1998/
_version_ 1848790701501841408
author Pal, Mahesh
Foody, Giles M.
author_facet Pal, Mahesh
Foody, Giles M.
author_sort Pal, Mahesh
building Nottingham Research Data Repository
collection Online Access
description SVM are attractive for the classification of remotely sensed data with some claims that the method is insensitive to the dimensionality of the data and so not requiring a dimensionality reduction analysis in pre-processing. Here, a series of classification analyses with two hyperspectral sensor data sets reveal that the accuracy of a classification by a SVM does vary as a function of the number of features used. Critically, it is shown that the accuracy of a classification may decline significantly (at 0.05 level of statistical significance) with the addition of features, especially if a small training sample is used. This highlights a dependency of the accuracy of classification by a SVM on the dimensionality of the data and so the potential value of undertaking a feature selection analysis prior to classification. Additionally, it is demonstrated that even when a large training sample is available feature selection may still be useful. For example, the accuracy derived from the use of a small number of features may be non-inferior (at 0.05% level of significance) to that derived from the use of a larger feature set providing potential advantages in relation to issues such as data storage and computational processing costs. Feature selection may, therefore, be a valuable analysis to include in pre-processing operations for classification by a SVM.
first_indexed 2025-11-14T18:16:48Z
format Article
id nottingham-1998
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T18:16:48Z
publishDate 2010
publisher Institute of Electrical and Electronics Engineers
recordtype eprints
repository_type Digital Repository
spelling nottingham-19982020-05-04T20:25:45Z https://eprints.nottingham.ac.uk/1998/ Feature selection for classification of hyperspectral data by SVM Pal, Mahesh Foody, Giles M. SVM are attractive for the classification of remotely sensed data with some claims that the method is insensitive to the dimensionality of the data and so not requiring a dimensionality reduction analysis in pre-processing. Here, a series of classification analyses with two hyperspectral sensor data sets reveal that the accuracy of a classification by a SVM does vary as a function of the number of features used. Critically, it is shown that the accuracy of a classification may decline significantly (at 0.05 level of statistical significance) with the addition of features, especially if a small training sample is used. This highlights a dependency of the accuracy of classification by a SVM on the dimensionality of the data and so the potential value of undertaking a feature selection analysis prior to classification. Additionally, it is demonstrated that even when a large training sample is available feature selection may still be useful. For example, the accuracy derived from the use of a small number of features may be non-inferior (at 0.05% level of significance) to that derived from the use of a larger feature set providing potential advantages in relation to issues such as data storage and computational processing costs. Feature selection may, therefore, be a valuable analysis to include in pre-processing operations for classification by a SVM. Institute of Electrical and Electronics Engineers 2010 Article PeerReviewed Pal, Mahesh and Foody, Giles M. (2010) Feature selection for classification of hyperspectral data by SVM. IEEE Transactions on Geoscience and Remote Sensing, 48 (5). pp. 2297-2307. ISSN 0196-2892 http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5419028 doi:10.1109/TGRS.2009.2039484 doi:10.1109/TGRS.2009.2039484
spellingShingle Pal, Mahesh
Foody, Giles M.
Feature selection for classification of hyperspectral data by SVM
title Feature selection for classification of hyperspectral data by SVM
title_full Feature selection for classification of hyperspectral data by SVM
title_fullStr Feature selection for classification of hyperspectral data by SVM
title_full_unstemmed Feature selection for classification of hyperspectral data by SVM
title_short Feature selection for classification of hyperspectral data by SVM
title_sort feature selection for classification of hyperspectral data by svm
url https://eprints.nottingham.ac.uk/1998/
https://eprints.nottingham.ac.uk/1998/
https://eprints.nottingham.ac.uk/1998/