A Dimension Reduction Framework for HSI Classification Using Fuzzy and Kernel NFLE Transformation
In this paper, a general nearest feature line (NFL) embedding (NFLE) transformation called fuzzy-kernel NFLE (FKNFLE) is proposed for hyperspectral image (HSI) classification in which kernelization and fuzzification are simultaneously considered. Though NFLE has successfully demonstrated its discrim...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2015-10-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | http://www.mdpi.com/2072-4292/7/11/14292 |
id |
doaj-b6cb5cfddbed4d4f8dde56b261d7a296 |
---|---|
record_format |
Article |
spelling |
doaj-b6cb5cfddbed4d4f8dde56b261d7a2962020-11-24T22:57:12ZengMDPI AGRemote Sensing2072-42922015-10-01711142921432610.3390/rs71114292rs71114292A Dimension Reduction Framework for HSI Classification Using Fuzzy and Kernel NFLE TransformationYing-Nong Chen0Cheng-Ta Hsieh1Ming-Gang Wen2Chin-Chuan Han3Kuo-Chin Fan4Department of Computer Science and Information Engineering, National Central University, Taoyuan 32001, TaiwanDepartment of Computer Science and Information Engineering, National Central University, Taoyuan 32001, TaiwanDepartment of Information Management, National United University, Miaoli 36063, TaiwanDepartment of Computer Science and Information Engineering, National United University, Miaoli 36063, TaiwanDepartment of Computer Science and Information Engineering, National Central University, Taoyuan 32001, TaiwanIn this paper, a general nearest feature line (NFL) embedding (NFLE) transformation called fuzzy-kernel NFLE (FKNFLE) is proposed for hyperspectral image (HSI) classification in which kernelization and fuzzification are simultaneously considered. Though NFLE has successfully demonstrated its discriminative capability, the non-linear manifold structure cannot be structured more efficiently by linear scatters using the linear NFLE method. According to the proposed scheme, samples were projected into a kernel space and assigned larger weights based on that of their neighbors. The within-class and between-class scatters were calculated using the fuzzy weights, and the best transformation was obtained by maximizing the Fisher criterion in the kernel space. In that way, the kernelized manifold learning preserved the local manifold structure in a Hilbert space as well as the locality of the manifold structure in the reduced low-dimensional space. The proposed method was compared with various state-of-the-art methods to evaluate the performance using three benchmark data sets. Based on the experimental results: the proposed FKNFLE outperformed the other, more conventional methods.http://www.mdpi.com/2072-4292/7/11/14292hyperspectral image classificationmanifold learningnearest feature line embeddingkernelizationfuzzification |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Ying-Nong Chen Cheng-Ta Hsieh Ming-Gang Wen Chin-Chuan Han Kuo-Chin Fan |
spellingShingle |
Ying-Nong Chen Cheng-Ta Hsieh Ming-Gang Wen Chin-Chuan Han Kuo-Chin Fan A Dimension Reduction Framework for HSI Classification Using Fuzzy and Kernel NFLE Transformation Remote Sensing hyperspectral image classification manifold learning nearest feature line embedding kernelization fuzzification |
author_facet |
Ying-Nong Chen Cheng-Ta Hsieh Ming-Gang Wen Chin-Chuan Han Kuo-Chin Fan |
author_sort |
Ying-Nong Chen |
title |
A Dimension Reduction Framework for HSI Classification Using Fuzzy and Kernel NFLE Transformation |
title_short |
A Dimension Reduction Framework for HSI Classification Using Fuzzy and Kernel NFLE Transformation |
title_full |
A Dimension Reduction Framework for HSI Classification Using Fuzzy and Kernel NFLE Transformation |
title_fullStr |
A Dimension Reduction Framework for HSI Classification Using Fuzzy and Kernel NFLE Transformation |
title_full_unstemmed |
A Dimension Reduction Framework for HSI Classification Using Fuzzy and Kernel NFLE Transformation |
title_sort |
dimension reduction framework for hsi classification using fuzzy and kernel nfle transformation |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2015-10-01 |
description |
In this paper, a general nearest feature line (NFL) embedding (NFLE) transformation called fuzzy-kernel NFLE (FKNFLE) is proposed for hyperspectral image (HSI) classification in which kernelization and fuzzification are simultaneously considered. Though NFLE has successfully demonstrated its discriminative capability, the non-linear manifold structure cannot be structured more efficiently by linear scatters using the linear NFLE method. According to the proposed scheme, samples were projected into a kernel space and assigned larger weights based on that of their neighbors. The within-class and between-class scatters were calculated using the fuzzy weights, and the best transformation was obtained by maximizing the Fisher criterion in the kernel space. In that way, the kernelized manifold learning preserved the local manifold structure in a Hilbert space as well as the locality of the manifold structure in the reduced low-dimensional space. The proposed method was compared with various state-of-the-art methods to evaluate the performance using three benchmark data sets. Based on the experimental results: the proposed FKNFLE outperformed the other, more conventional methods. |
topic |
hyperspectral image classification manifold learning nearest feature line embedding kernelization fuzzification |
url |
http://www.mdpi.com/2072-4292/7/11/14292 |
work_keys_str_mv |
AT yingnongchen adimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT chengtahsieh adimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT minggangwen adimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT chinchuanhan adimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT kuochinfan adimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT yingnongchen dimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT chengtahsieh dimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT minggangwen dimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT chinchuanhan dimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation AT kuochinfan dimensionreductionframeworkforhsiclassificationusingfuzzyandkernelnfletransformation |
_version_ |
1725651315490553856 |