Decorrelation-Separability-Based Affinity Propagation for Semisupervised Clustering of Hyperspectral Images
Dimensionality reduction is a common approach to decrease the high computational complexity and improve the classification performance of hyperspectral images. The paper addresses this issue by introducing a novel semisupervised clustering approach to hyperspectral image classification. In this appr...
Published in: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
country:USA
2016
|
Subjects: | |
Online Access: | http://hdl.handle.net/11572/168544 https://doi.org/10.1109/JSTARS.2015.2461658 https://ieeexplore.ieee.org/document/7185332 |
id |
ftutrentoiris:oai:iris.unitn.it:11572/168544 |
---|---|
record_format |
openpolar |
spelling |
ftutrentoiris:oai:iris.unitn.it:11572/168544 2024-02-11T10:03:23+01:00 Decorrelation-Separability-Based Affinity Propagation for Semisupervised Clustering of Hyperspectral Images Yang, Chen Bruzzone, Lorenzo Zhao, Haishi Liang, Yanchun Guan, Renchu Yang, Chen Bruzzone, Lorenzo Zhao, Haishi Liang, Yanchun Guan, Renchu 2016 http://hdl.handle.net/11572/168544 https://doi.org/10.1109/JSTARS.2015.2461658 https://ieeexplore.ieee.org/document/7185332 eng eng country:USA info:eu-repo/semantics/altIdentifier/wos/WOS:000370877600003 volume:9 issue:2 firstpage:568 lastpage:582 numberofpages:15 journal:IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING http://hdl.handle.net/11572/168544 doi:10.1109/JSTARS.2015.2461658 info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-84939134414 https://ieeexplore.ieee.org/document/7185332 info:eu-repo/semantics/closedAccess Affinity propagation (AP) clustering discriminative component analysis (DCA) distance metric learning (DML) hyperspectral image remote sensing Computers in Earth Science Atmospheric Science info:eu-repo/semantics/article 2016 ftutrentoiris https://doi.org/10.1109/JSTARS.2015.2461658 2024-01-23T23:09:35Z Dimensionality reduction is a common approach to decrease the high computational complexity and improve the classification performance of hyperspectral images. The paper addresses this issue by introducing a novel semisupervised clustering approach to hyperspectral image classification. In this approach, the relationships between the samples (i.e., pixels in hyperspectral data) are measured by two kinds of side constraints, i.e., positive and negative constraints, which are imposed to construct a discriminative transformation that establishes a regularized metric function. Accordingly, a new subspace is built in which the class discrimination capability of each individual feature is expanded, while the spectral correlation among features is greatly reduced. Then, the learned metric is formulated within an exemplar-based clustering framework, i.e., the affinity propagation (AP). Thus, the proposed approach is called decorrelation-separability-based AP (DS-AP). Experimental results obtained on three hyperspectral remote sensing data sets demonstrate the effectiveness of the proposed DS-AP technique for hyperspectral image classification. Article in Journal/Newspaper DML Università degli Studi di Trento: CINECA IRIS IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 9 2 568 582 |
institution |
Open Polar |
collection |
Università degli Studi di Trento: CINECA IRIS |
op_collection_id |
ftutrentoiris |
language |
English |
topic |
Affinity propagation (AP) clustering discriminative component analysis (DCA) distance metric learning (DML) hyperspectral image remote sensing Computers in Earth Science Atmospheric Science |
spellingShingle |
Affinity propagation (AP) clustering discriminative component analysis (DCA) distance metric learning (DML) hyperspectral image remote sensing Computers in Earth Science Atmospheric Science Yang, Chen Bruzzone, Lorenzo Zhao, Haishi Liang, Yanchun Guan, Renchu Decorrelation-Separability-Based Affinity Propagation for Semisupervised Clustering of Hyperspectral Images |
topic_facet |
Affinity propagation (AP) clustering discriminative component analysis (DCA) distance metric learning (DML) hyperspectral image remote sensing Computers in Earth Science Atmospheric Science |
description |
Dimensionality reduction is a common approach to decrease the high computational complexity and improve the classification performance of hyperspectral images. The paper addresses this issue by introducing a novel semisupervised clustering approach to hyperspectral image classification. In this approach, the relationships between the samples (i.e., pixels in hyperspectral data) are measured by two kinds of side constraints, i.e., positive and negative constraints, which are imposed to construct a discriminative transformation that establishes a regularized metric function. Accordingly, a new subspace is built in which the class discrimination capability of each individual feature is expanded, while the spectral correlation among features is greatly reduced. Then, the learned metric is formulated within an exemplar-based clustering framework, i.e., the affinity propagation (AP). Thus, the proposed approach is called decorrelation-separability-based AP (DS-AP). Experimental results obtained on three hyperspectral remote sensing data sets demonstrate the effectiveness of the proposed DS-AP technique for hyperspectral image classification. |
author2 |
Yang, Chen Bruzzone, Lorenzo Zhao, Haishi Liang, Yanchun Guan, Renchu |
format |
Article in Journal/Newspaper |
author |
Yang, Chen Bruzzone, Lorenzo Zhao, Haishi Liang, Yanchun Guan, Renchu |
author_facet |
Yang, Chen Bruzzone, Lorenzo Zhao, Haishi Liang, Yanchun Guan, Renchu |
author_sort |
Yang, Chen |
title |
Decorrelation-Separability-Based Affinity Propagation for Semisupervised Clustering of Hyperspectral Images |
title_short |
Decorrelation-Separability-Based Affinity Propagation for Semisupervised Clustering of Hyperspectral Images |
title_full |
Decorrelation-Separability-Based Affinity Propagation for Semisupervised Clustering of Hyperspectral Images |
title_fullStr |
Decorrelation-Separability-Based Affinity Propagation for Semisupervised Clustering of Hyperspectral Images |
title_full_unstemmed |
Decorrelation-Separability-Based Affinity Propagation for Semisupervised Clustering of Hyperspectral Images |
title_sort |
decorrelation-separability-based affinity propagation for semisupervised clustering of hyperspectral images |
publisher |
country:USA |
publishDate |
2016 |
url |
http://hdl.handle.net/11572/168544 https://doi.org/10.1109/JSTARS.2015.2461658 https://ieeexplore.ieee.org/document/7185332 |
genre |
DML |
genre_facet |
DML |
op_relation |
info:eu-repo/semantics/altIdentifier/wos/WOS:000370877600003 volume:9 issue:2 firstpage:568 lastpage:582 numberofpages:15 journal:IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING http://hdl.handle.net/11572/168544 doi:10.1109/JSTARS.2015.2461658 info:eu-repo/semantics/altIdentifier/scopus/2-s2.0-84939134414 https://ieeexplore.ieee.org/document/7185332 |
op_rights |
info:eu-repo/semantics/closedAccess |
op_doi |
https://doi.org/10.1109/JSTARS.2015.2461658 |
container_title |
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
container_volume |
9 |
container_issue |
2 |
container_start_page |
568 |
op_container_end_page |
582 |
_version_ |
1790599599378923520 |