Tensor Based Multiscale Low Rank Decomposition for Hyperspectral Images Dimensionality Reduction

Dimensionality reduction is an essential and important issue in hyperspectral image processing. With the advantages of preserving the spatial neighborhood information and the global structure information, tensor analysis and low rank representation have been widely considered in this field and yield...

Full description

Bibliographic Details
Main Authors: Jinliang An, Jinhui Lei, Yuzhen Song, Xiangrong Zhang, Jinmei Guo
Format: Article
Language:English
Published: MDPI AG 2019-06-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/11/12/1485
id doaj-c2f2b00f34194affb0957fbc46100317
record_format Article
spelling doaj-c2f2b00f34194affb0957fbc461003172020-11-25T01:09:00ZengMDPI AGRemote Sensing2072-42922019-06-011112148510.3390/rs11121485rs11121485Tensor Based Multiscale Low Rank Decomposition for Hyperspectral Images Dimensionality ReductionJinliang An0Jinhui Lei1Yuzhen Song2Xiangrong Zhang3Jinmei Guo4School of Information Engineering, Henan Institute of Science and Technology, Xinxiang 453003, ChinaSchool of Information Engineering, Henan Institute of Science and Technology, Xinxiang 453003, ChinaSchool of Information Engineering, Henan Institute of Science and Technology, Xinxiang 453003, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaSchool of Artificial Intelligence, Xidian University, Xi’an 710071, ChinaDimensionality reduction is an essential and important issue in hyperspectral image processing. With the advantages of preserving the spatial neighborhood information and the global structure information, tensor analysis and low rank representation have been widely considered in this field and yielded satisfactory performance. In available tensor- and low rank-based methods, how to construct appropriate tensor samples and determine the optimal rank of hyperspectral images along each mode are still challenging issues. To address these drawbacks, an unsupervised tensor-based multiscale low rank decomposition (T-MLRD) method for hyperspectral images dimensionality reduction is proposed in this paper. By regarding the raw cube hyperspectral image as the only tensor sample, T-MLRD needs no labeled samples and avoids the processing of constructing tensor samples. In addition, a novel multiscale low rank estimating method is proposed to obtain the optimal rank along each mode of hyperspectral image which avoids the complicated rank computing. Finally, the multiscale low rank feature representation is fused to achieve dimensionality reduction. Experimental results on real hyperspectral datasets demonstrate the superiority of the proposed method over several state-of-the-art approaches.https://www.mdpi.com/2072-4292/11/12/1485dimensionality reductionhyperspectral images classificationmultiscalelow rank
collection DOAJ
language English
format Article
sources DOAJ
author Jinliang An
Jinhui Lei
Yuzhen Song
Xiangrong Zhang
Jinmei Guo
spellingShingle Jinliang An
Jinhui Lei
Yuzhen Song
Xiangrong Zhang
Jinmei Guo
Tensor Based Multiscale Low Rank Decomposition for Hyperspectral Images Dimensionality Reduction
Remote Sensing
dimensionality reduction
hyperspectral images classification
multiscale
low rank
author_facet Jinliang An
Jinhui Lei
Yuzhen Song
Xiangrong Zhang
Jinmei Guo
author_sort Jinliang An
title Tensor Based Multiscale Low Rank Decomposition for Hyperspectral Images Dimensionality Reduction
title_short Tensor Based Multiscale Low Rank Decomposition for Hyperspectral Images Dimensionality Reduction
title_full Tensor Based Multiscale Low Rank Decomposition for Hyperspectral Images Dimensionality Reduction
title_fullStr Tensor Based Multiscale Low Rank Decomposition for Hyperspectral Images Dimensionality Reduction
title_full_unstemmed Tensor Based Multiscale Low Rank Decomposition for Hyperspectral Images Dimensionality Reduction
title_sort tensor based multiscale low rank decomposition for hyperspectral images dimensionality reduction
publisher MDPI AG
series Remote Sensing
issn 2072-4292
publishDate 2019-06-01
description Dimensionality reduction is an essential and important issue in hyperspectral image processing. With the advantages of preserving the spatial neighborhood information and the global structure information, tensor analysis and low rank representation have been widely considered in this field and yielded satisfactory performance. In available tensor- and low rank-based methods, how to construct appropriate tensor samples and determine the optimal rank of hyperspectral images along each mode are still challenging issues. To address these drawbacks, an unsupervised tensor-based multiscale low rank decomposition (T-MLRD) method for hyperspectral images dimensionality reduction is proposed in this paper. By regarding the raw cube hyperspectral image as the only tensor sample, T-MLRD needs no labeled samples and avoids the processing of constructing tensor samples. In addition, a novel multiscale low rank estimating method is proposed to obtain the optimal rank along each mode of hyperspectral image which avoids the complicated rank computing. Finally, the multiscale low rank feature representation is fused to achieve dimensionality reduction. Experimental results on real hyperspectral datasets demonstrate the superiority of the proposed method over several state-of-the-art approaches.
topic dimensionality reduction
hyperspectral images classification
multiscale
low rank
url https://www.mdpi.com/2072-4292/11/12/1485
work_keys_str_mv AT jinliangan tensorbasedmultiscalelowrankdecompositionforhyperspectralimagesdimensionalityreduction
AT jinhuilei tensorbasedmultiscalelowrankdecompositionforhyperspectralimagesdimensionalityreduction
AT yuzhensong tensorbasedmultiscalelowrankdecompositionforhyperspectralimagesdimensionalityreduction
AT xiangrongzhang tensorbasedmultiscalelowrankdecompositionforhyperspectralimagesdimensionalityreduction
AT jinmeiguo tensorbasedmultiscalelowrankdecompositionforhyperspectralimagesdimensionalityreduction
_version_ 1725180477166321664