Geometrical Approximated Principal Component Analysis for Hyperspectral Image Analysis
Principal Component Analysis (PCA) is a method based on statistics and linear algebra techniques, used in hyperspectral satellite imagery for data dimensionality reduction required in order to speed up and increase the performance of subsequent hyperspectral image processing algorithms. This paper i...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-05-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/12/11/1698 |
_version_ | 1797566990090502144 |
---|---|
author | Alina L. Machidon Fabio Del Frate Matteo Picchiani Octavian M. Machidon Petre L. Ogrutan |
author_facet | Alina L. Machidon Fabio Del Frate Matteo Picchiani Octavian M. Machidon Petre L. Ogrutan |
author_sort | Alina L. Machidon |
collection | DOAJ |
description | Principal Component Analysis (PCA) is a method based on statistics and linear algebra techniques, used in hyperspectral satellite imagery for data dimensionality reduction required in order to speed up and increase the performance of subsequent hyperspectral image processing algorithms. This paper introduces the PCA approximation method based on a geometric construction approach (gaPCA) method, an alternative algorithm for computing the principal components based on a geometrical constructed approximation of the standard PCA and presents its application to remote sensing hyperspectral images. gaPCA has the potential of yielding better land classification results by preserving a higher degree of information related to the smaller objects of the scene (or to the rare spectral objects) than the standard PCA, being focused not on maximizing the variance of the data, but the range. The paper validates gaPCA on four distinct datasets and performs comparative evaluations and metrics with the standard PCA method. A comparative land classification benchmark of gaPCA and the standard PCA using statistical-based tools is also described. The results show gaPCA is an effective dimensionality-reduction tool, with performance similar to, and in several cases, even higher than standard PCA on specific image classification tasks. gaPCA was shown to be more suitable for hyperspectral images with small structures or objects that need to be detected or where preponderantly spectral classes or spectrally similar classes are present. |
first_indexed | 2024-03-10T19:34:44Z |
format | Article |
id | doaj.art-71e497b544cf4d53a89987416ba64bfa |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-10T19:34:44Z |
publishDate | 2020-05-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-71e497b544cf4d53a89987416ba64bfa2023-11-20T01:46:47ZengMDPI AGRemote Sensing2072-42922020-05-011211169810.3390/rs12111698Geometrical Approximated Principal Component Analysis for Hyperspectral Image AnalysisAlina L. Machidon0Fabio Del Frate1Matteo Picchiani2Octavian M. Machidon3Petre L. Ogrutan4Department of Electronics and Computers, Transilvania University of Brasov, 500036 Brasov, RomaniaDepartment of Civil Engineering and Computer Science Engineering, University of “Tor Vergata”, 00133 Rome, ItalyGEO-K s.r.l, 00133 Rome, ItalyDepartment of Electronics and Computers, Transilvania University of Brasov, 500036 Brasov, RomaniaDepartment of Electronics and Computers, Transilvania University of Brasov, 500036 Brasov, RomaniaPrincipal Component Analysis (PCA) is a method based on statistics and linear algebra techniques, used in hyperspectral satellite imagery for data dimensionality reduction required in order to speed up and increase the performance of subsequent hyperspectral image processing algorithms. This paper introduces the PCA approximation method based on a geometric construction approach (gaPCA) method, an alternative algorithm for computing the principal components based on a geometrical constructed approximation of the standard PCA and presents its application to remote sensing hyperspectral images. gaPCA has the potential of yielding better land classification results by preserving a higher degree of information related to the smaller objects of the scene (or to the rare spectral objects) than the standard PCA, being focused not on maximizing the variance of the data, but the range. The paper validates gaPCA on four distinct datasets and performs comparative evaluations and metrics with the standard PCA method. A comparative land classification benchmark of gaPCA and the standard PCA using statistical-based tools is also described. The results show gaPCA is an effective dimensionality-reduction tool, with performance similar to, and in several cases, even higher than standard PCA on specific image classification tasks. gaPCA was shown to be more suitable for hyperspectral images with small structures or objects that need to be detected or where preponderantly spectral classes or spectrally similar classes are present.https://www.mdpi.com/2072-4292/12/11/1698hyperspectral imagedimensionality reductionprincipal component analysisland classification |
spellingShingle | Alina L. Machidon Fabio Del Frate Matteo Picchiani Octavian M. Machidon Petre L. Ogrutan Geometrical Approximated Principal Component Analysis for Hyperspectral Image Analysis Remote Sensing hyperspectral image dimensionality reduction principal component analysis land classification |
title | Geometrical Approximated Principal Component Analysis for Hyperspectral Image Analysis |
title_full | Geometrical Approximated Principal Component Analysis for Hyperspectral Image Analysis |
title_fullStr | Geometrical Approximated Principal Component Analysis for Hyperspectral Image Analysis |
title_full_unstemmed | Geometrical Approximated Principal Component Analysis for Hyperspectral Image Analysis |
title_short | Geometrical Approximated Principal Component Analysis for Hyperspectral Image Analysis |
title_sort | geometrical approximated principal component analysis for hyperspectral image analysis |
topic | hyperspectral image dimensionality reduction principal component analysis land classification |
url | https://www.mdpi.com/2072-4292/12/11/1698 |
work_keys_str_mv | AT alinalmachidon geometricalapproximatedprincipalcomponentanalysisforhyperspectralimageanalysis AT fabiodelfrate geometricalapproximatedprincipalcomponentanalysisforhyperspectralimageanalysis AT matteopicchiani geometricalapproximatedprincipalcomponentanalysisforhyperspectralimageanalysis AT octavianmmachidon geometricalapproximatedprincipalcomponentanalysisforhyperspectralimageanalysis AT petrelogrutan geometricalapproximatedprincipalcomponentanalysisforhyperspectralimageanalysis |