Discovering the Representative Subset with Low Redundancy for Hyperspectral Feature Selection
In this paper, a novel unsupervised band selection (BS) criterion based on maximizing representativeness and minimizing redundancy (MRMR) is proposed for selecting a set of informative bands to represent the whole hyperspectral image cube. The new selection criterion is denoted as the MRMR selection...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2019-06-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/11/11/1341 |
id |
doaj-dda4f2961b2d427cb1ad24aea0b72a4e |
---|---|
record_format |
Article |
spelling |
doaj-dda4f2961b2d427cb1ad24aea0b72a4e2020-11-24T21:47:41ZengMDPI AGRemote Sensing2072-42922019-06-011111134110.3390/rs11111341rs11111341Discovering the Representative Subset with Low Redundancy for Hyperspectral Feature SelectionWenqiang Zhang0Xiaorun Li1Liaoying Zhao2College of Electrical Engineering, Zhejiang University, No.38, Zheda Road, Xihu District, Hangzhou 310027, ChinaCollege of Electrical Engineering, Zhejiang University, No.38, Zheda Road, Xihu District, Hangzhou 310027, ChinaDepartment of Computer Science, Hangzhou Dianzi University, Zhejiang 310027, ChinaIn this paper, a novel unsupervised band selection (BS) criterion based on maximizing representativeness and minimizing redundancy (MRMR) is proposed for selecting a set of informative bands to represent the whole hyperspectral image cube. The new selection criterion is denoted as the MRMR selection criterion and the associated BS method is denoted as the MRMR method. The MRMR selection criterion can evaluate the band subset’s representativeness and redundancy simultaneously. For one band subset, its representativeness is estimated by using orthogonal projection (OP) and its redundancy is measured by the average of the Pearson correlation coefficients among the bands in this subset. To find the satisfactory subset, an effective evolutionary algorithm, i.e., the immune clone selection (ICS) algorithm, is applied as the subset searching strategy. Moreover, we further introduce two effective tricks to simplify the computation of the representativeness metric, thus the computational complexity of the proposed method is reduced significantly. Experimental results on different real-world datasets demonstrate that the proposed method is very effective and its selected bands can obtain good classification performances in practice.https://www.mdpi.com/2072-4292/11/11/1341unsupervised feature selectiondimensionality reductionhyperspectral imageorthogonal projectionevolutionary algorithm |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Wenqiang Zhang Xiaorun Li Liaoying Zhao |
spellingShingle |
Wenqiang Zhang Xiaorun Li Liaoying Zhao Discovering the Representative Subset with Low Redundancy for Hyperspectral Feature Selection Remote Sensing unsupervised feature selection dimensionality reduction hyperspectral image orthogonal projection evolutionary algorithm |
author_facet |
Wenqiang Zhang Xiaorun Li Liaoying Zhao |
author_sort |
Wenqiang Zhang |
title |
Discovering the Representative Subset with Low Redundancy for Hyperspectral Feature Selection |
title_short |
Discovering the Representative Subset with Low Redundancy for Hyperspectral Feature Selection |
title_full |
Discovering the Representative Subset with Low Redundancy for Hyperspectral Feature Selection |
title_fullStr |
Discovering the Representative Subset with Low Redundancy for Hyperspectral Feature Selection |
title_full_unstemmed |
Discovering the Representative Subset with Low Redundancy for Hyperspectral Feature Selection |
title_sort |
discovering the representative subset with low redundancy for hyperspectral feature selection |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2019-06-01 |
description |
In this paper, a novel unsupervised band selection (BS) criterion based on maximizing representativeness and minimizing redundancy (MRMR) is proposed for selecting a set of informative bands to represent the whole hyperspectral image cube. The new selection criterion is denoted as the MRMR selection criterion and the associated BS method is denoted as the MRMR method. The MRMR selection criterion can evaluate the band subset’s representativeness and redundancy simultaneously. For one band subset, its representativeness is estimated by using orthogonal projection (OP) and its redundancy is measured by the average of the Pearson correlation coefficients among the bands in this subset. To find the satisfactory subset, an effective evolutionary algorithm, i.e., the immune clone selection (ICS) algorithm, is applied as the subset searching strategy. Moreover, we further introduce two effective tricks to simplify the computation of the representativeness metric, thus the computational complexity of the proposed method is reduced significantly. Experimental results on different real-world datasets demonstrate that the proposed method is very effective and its selected bands can obtain good classification performances in practice. |
topic |
unsupervised feature selection dimensionality reduction hyperspectral image orthogonal projection evolutionary algorithm |
url |
https://www.mdpi.com/2072-4292/11/11/1341 |
work_keys_str_mv |
AT wenqiangzhang discoveringtherepresentativesubsetwithlowredundancyforhyperspectralfeatureselection AT xiaorunli discoveringtherepresentativesubsetwithlowredundancyforhyperspectralfeatureselection AT liaoyingzhao discoveringtherepresentativesubsetwithlowredundancyforhyperspectralfeatureselection |
_version_ |
1725896309227913216 |