Band Priority Index: A Feature Selection Framework for Hyperspectral Imagery
Hyperspectral Band Selection (BS) aims to select a few informative and distinctive bands to represent the whole image cube. In this paper, an unsupervised BS framework named the band priority index (BPI) is proposed. The basic idea of BPI is to find the bands with large amounts of information and lo...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2018-07-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | http://www.mdpi.com/2072-4292/10/7/1095 |
id |
doaj-3cb5040cc1ed4ea0844d23c329ad8a71 |
---|---|
record_format |
Article |
spelling |
doaj-3cb5040cc1ed4ea0844d23c329ad8a712020-11-24T21:20:11ZengMDPI AGRemote Sensing2072-42922018-07-01107109510.3390/rs10071095rs10071095Band Priority Index: A Feature Selection Framework for Hyperspectral ImageryWenqiang 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, ChinaHyperspectral Band Selection (BS) aims to select a few informative and distinctive bands to represent the whole image cube. In this paper, an unsupervised BS framework named the band priority index (BPI) is proposed. The basic idea of BPI is to find the bands with large amounts of information and low correlation. Sequential forward search (SFS) is used to avoid an exhaustive search, and the objective function of BPI consist of two parts: the information metric and the correlation metric. We proposed a new band correlation metric, namely, the joint correlation coefficient (JCC), to estimate the joint correlation between a single band and multiple bands. JCC uses the angle between a band and the hyperplane determined by a band set to evaluate the correlation between them. To estimate the amount of information, the variance and entropy are used as the information metric for BPI, respectively. Since BPI is a framework for BS, other information metrics and different mathematic functions of the angle can also be used in the model, which means there are various implementations of BPI. The BPI-based methods have the advantages as follows: (1) The selected bands are informative and distinctive. (2) The BPI-based methods usually have good computational efficiencies. (3) These methods have the potential to determine the number of bands to be selected. The experimental results on different real hyperspectral datasets demonstrate that the BPI-based methods are highly efficient and accurate BS methods.http://www.mdpi.com/2072-4292/10/7/1095feature selectiondimensionality reductionhyperspectral remote sensingclassificationgreedy search algorithm |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Wenqiang Zhang Xiaorun Li Liaoying Zhao |
spellingShingle |
Wenqiang Zhang Xiaorun Li Liaoying Zhao Band Priority Index: A Feature Selection Framework for Hyperspectral Imagery Remote Sensing feature selection dimensionality reduction hyperspectral remote sensing classification greedy search algorithm |
author_facet |
Wenqiang Zhang Xiaorun Li Liaoying Zhao |
author_sort |
Wenqiang Zhang |
title |
Band Priority Index: A Feature Selection Framework for Hyperspectral Imagery |
title_short |
Band Priority Index: A Feature Selection Framework for Hyperspectral Imagery |
title_full |
Band Priority Index: A Feature Selection Framework for Hyperspectral Imagery |
title_fullStr |
Band Priority Index: A Feature Selection Framework for Hyperspectral Imagery |
title_full_unstemmed |
Band Priority Index: A Feature Selection Framework for Hyperspectral Imagery |
title_sort |
band priority index: a feature selection framework for hyperspectral imagery |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2018-07-01 |
description |
Hyperspectral Band Selection (BS) aims to select a few informative and distinctive bands to represent the whole image cube. In this paper, an unsupervised BS framework named the band priority index (BPI) is proposed. The basic idea of BPI is to find the bands with large amounts of information and low correlation. Sequential forward search (SFS) is used to avoid an exhaustive search, and the objective function of BPI consist of two parts: the information metric and the correlation metric. We proposed a new band correlation metric, namely, the joint correlation coefficient (JCC), to estimate the joint correlation between a single band and multiple bands. JCC uses the angle between a band and the hyperplane determined by a band set to evaluate the correlation between them. To estimate the amount of information, the variance and entropy are used as the information metric for BPI, respectively. Since BPI is a framework for BS, other information metrics and different mathematic functions of the angle can also be used in the model, which means there are various implementations of BPI. The BPI-based methods have the advantages as follows: (1) The selected bands are informative and distinctive. (2) The BPI-based methods usually have good computational efficiencies. (3) These methods have the potential to determine the number of bands to be selected. The experimental results on different real hyperspectral datasets demonstrate that the BPI-based methods are highly efficient and accurate BS methods. |
topic |
feature selection dimensionality reduction hyperspectral remote sensing classification greedy search algorithm |
url |
http://www.mdpi.com/2072-4292/10/7/1095 |
work_keys_str_mv |
AT wenqiangzhang bandpriorityindexafeatureselectionframeworkforhyperspectralimagery AT xiaorunli bandpriorityindexafeatureselectionframeworkforhyperspectralimagery AT liaoyingzhao bandpriorityindexafeatureselectionframeworkforhyperspectralimagery |
_version_ |
1726003486110253056 |