Novel two dimensional singular spectrum analysis for effective feature extraction and data classification in hyperspectral imaging

Jaime Zabalza, Jinchang Ren, Jiangbin Zheng, Junwei Han, Huimin Zhao, Shutao Li, Stephen Marshall

Research output: Contribution to journalArticle

68 Citations (Scopus)

Abstract

Feature extraction is of high importance for effective data classification in hyperspectral imaging (HSI). Considering the high correlation among band images, spectral-domain feature extraction is widely employed. For effective spatial information extraction, a 2-D extension to singular spectrum analysis (SSA), a recent technique for generic data mining and temporal signal analysis, is proposed. With 2D-SSA applied to HSI, each band image is decomposed into varying trend, oscillations and noise. Using the trend and selected oscillations as features, the reconstructed signal, with noise highly suppressed, becomes more robust and effective for data classification. Three publicly available data sets for HSI remote sensing data classification are used in our experiments. Comprehensive results using a support vector machine (SVM) classifier have quantitatively evaluated the efficacy of the proposed approach. Benchmarked with several state-of-the-art methods including 2-D empirical mode decomposition (2D-EMD), it is found that our proposed 2D-SSA approach generates the best results in most cases. Unlike 2D-EMD which requires sequential transforms to obtain detailed decomposition, 2D-SSA extracts all components simultaneously. As a result, the executive time in feature extraction can also be dramatically reduced. The superiority in terms of enhanced discrimination ability from 2D-SSA is further validated when a relatively weak classifier, k-nearest neighbor (k-NN), is used for data classification. In addition, the combination of 2D-SSA with 1D-PCA (2D-SSA-PCA) has generated the best results among several other approaches, which has demonstrated the great potential in combining 2D-SSA with other approaches for effective spatial-spectral feature extraction and dimension reduction in HSI.
LanguageEnglish
Pages4418-4433
Number of pages16
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume53
Issue number8
Early online date20 Feb 2015
DOIs
Publication statusPublished - 31 Aug 2015

Fingerprint

Spectrum analysis
Feature extraction
Decomposition
decomposition
Classifiers
oscillation
Signal analysis
Hyperspectral imaging
analysis
data mining
Support vector machines
Data mining
Remote sensing
transform
remote sensing
Experiments
experiment

Keywords

  • data classification
  • feature extraction
  • hyperspectral imaging
  • 2-D empirical mode decomposition
  • 2-D singular spectrum analysis

Cite this

@article{781c4c9f3d7a413986deb5e2c586ce4d,
title = "Novel two dimensional singular spectrum analysis for effective feature extraction and data classification in hyperspectral imaging",
abstract = "Feature extraction is of high importance for effective data classification in hyperspectral imaging (HSI). Considering the high correlation among band images, spectral-domain feature extraction is widely employed. For effective spatial information extraction, a 2-D extension to singular spectrum analysis (SSA), a recent technique for generic data mining and temporal signal analysis, is proposed. With 2D-SSA applied to HSI, each band image is decomposed into varying trend, oscillations and noise. Using the trend and selected oscillations as features, the reconstructed signal, with noise highly suppressed, becomes more robust and effective for data classification. Three publicly available data sets for HSI remote sensing data classification are used in our experiments. Comprehensive results using a support vector machine (SVM) classifier have quantitatively evaluated the efficacy of the proposed approach. Benchmarked with several state-of-the-art methods including 2-D empirical mode decomposition (2D-EMD), it is found that our proposed 2D-SSA approach generates the best results in most cases. Unlike 2D-EMD which requires sequential transforms to obtain detailed decomposition, 2D-SSA extracts all components simultaneously. As a result, the executive time in feature extraction can also be dramatically reduced. The superiority in terms of enhanced discrimination ability from 2D-SSA is further validated when a relatively weak classifier, k-nearest neighbor (k-NN), is used for data classification. In addition, the combination of 2D-SSA with 1D-PCA (2D-SSA-PCA) has generated the best results among several other approaches, which has demonstrated the great potential in combining 2D-SSA with other approaches for effective spatial-spectral feature extraction and dimension reduction in HSI.",
keywords = "data classification, feature extraction, hyperspectral imaging, 2-D empirical mode decomposition, 2-D singular spectrum analysis",
author = "Jaime Zabalza and Jinchang Ren and Jiangbin Zheng and Junwei Han and Huimin Zhao and Shutao Li and Stephen Marshall",
note = "(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works",
year = "2015",
month = "8",
day = "31",
doi = "10.1109/TGRS.2015.2398468",
language = "English",
volume = "53",
pages = "4418--4433",
journal = "IEEE Transactions on Geoscience and Remote Sensing",
issn = "0196-2892",
publisher = "IEEE",
number = "8",

}

Novel two dimensional singular spectrum analysis for effective feature extraction and data classification in hyperspectral imaging. / Zabalza, Jaime; Ren, Jinchang; Zheng, Jiangbin; Han, Junwei; Zhao, Huimin; Li, Shutao; Marshall, Stephen.

In: IEEE Transactions on Geoscience and Remote Sensing, Vol. 53, No. 8, 31.08.2015, p. 4418-4433.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Novel two dimensional singular spectrum analysis for effective feature extraction and data classification in hyperspectral imaging

AU - Zabalza, Jaime

AU - Ren, Jinchang

AU - Zheng, Jiangbin

AU - Han, Junwei

AU - Zhao, Huimin

AU - Li, Shutao

AU - Marshall, Stephen

N1 - (c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works

PY - 2015/8/31

Y1 - 2015/8/31

N2 - Feature extraction is of high importance for effective data classification in hyperspectral imaging (HSI). Considering the high correlation among band images, spectral-domain feature extraction is widely employed. For effective spatial information extraction, a 2-D extension to singular spectrum analysis (SSA), a recent technique for generic data mining and temporal signal analysis, is proposed. With 2D-SSA applied to HSI, each band image is decomposed into varying trend, oscillations and noise. Using the trend and selected oscillations as features, the reconstructed signal, with noise highly suppressed, becomes more robust and effective for data classification. Three publicly available data sets for HSI remote sensing data classification are used in our experiments. Comprehensive results using a support vector machine (SVM) classifier have quantitatively evaluated the efficacy of the proposed approach. Benchmarked with several state-of-the-art methods including 2-D empirical mode decomposition (2D-EMD), it is found that our proposed 2D-SSA approach generates the best results in most cases. Unlike 2D-EMD which requires sequential transforms to obtain detailed decomposition, 2D-SSA extracts all components simultaneously. As a result, the executive time in feature extraction can also be dramatically reduced. The superiority in terms of enhanced discrimination ability from 2D-SSA is further validated when a relatively weak classifier, k-nearest neighbor (k-NN), is used for data classification. In addition, the combination of 2D-SSA with 1D-PCA (2D-SSA-PCA) has generated the best results among several other approaches, which has demonstrated the great potential in combining 2D-SSA with other approaches for effective spatial-spectral feature extraction and dimension reduction in HSI.

AB - Feature extraction is of high importance for effective data classification in hyperspectral imaging (HSI). Considering the high correlation among band images, spectral-domain feature extraction is widely employed. For effective spatial information extraction, a 2-D extension to singular spectrum analysis (SSA), a recent technique for generic data mining and temporal signal analysis, is proposed. With 2D-SSA applied to HSI, each band image is decomposed into varying trend, oscillations and noise. Using the trend and selected oscillations as features, the reconstructed signal, with noise highly suppressed, becomes more robust and effective for data classification. Three publicly available data sets for HSI remote sensing data classification are used in our experiments. Comprehensive results using a support vector machine (SVM) classifier have quantitatively evaluated the efficacy of the proposed approach. Benchmarked with several state-of-the-art methods including 2-D empirical mode decomposition (2D-EMD), it is found that our proposed 2D-SSA approach generates the best results in most cases. Unlike 2D-EMD which requires sequential transforms to obtain detailed decomposition, 2D-SSA extracts all components simultaneously. As a result, the executive time in feature extraction can also be dramatically reduced. The superiority in terms of enhanced discrimination ability from 2D-SSA is further validated when a relatively weak classifier, k-nearest neighbor (k-NN), is used for data classification. In addition, the combination of 2D-SSA with 1D-PCA (2D-SSA-PCA) has generated the best results among several other approaches, which has demonstrated the great potential in combining 2D-SSA with other approaches for effective spatial-spectral feature extraction and dimension reduction in HSI.

KW - data classification

KW - feature extraction

KW - hyperspectral imaging

KW - 2-D empirical mode decomposition

KW - 2-D singular spectrum analysis

U2 - 10.1109/TGRS.2015.2398468

DO - 10.1109/TGRS.2015.2398468

M3 - Article

VL - 53

SP - 4418

EP - 4433

JO - IEEE Transactions on Geoscience and Remote Sensing

T2 - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

SN - 0196-2892

IS - 8

ER -