Unsupervised Identification of Targeted Spectra Applying Rank<sup>1</sup>-NMF and FCC Algorithms in Long-Wave Hyperspectral Infrared Imagery
Clustering methods unequivocally show considerable influence on many recent algorithms and play an important role in hyperspectral data analysis. Here, we challenge the clustering for mineral identification using two different strategies in hyperspectral long wave infrared (LWIR, 7.7–11.8 μm). For t...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-05-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/13/11/2125 |
id |
doaj-88a02cf270fc4b93aff3baf23dccd87a |
---|---|
record_format |
Article |
spelling |
doaj-88a02cf270fc4b93aff3baf23dccd87a2021-06-01T01:30:00ZengMDPI AGRemote Sensing2072-42922021-05-01132125212510.3390/rs13112125Unsupervised Identification of Targeted Spectra Applying Rank<sup>1</sup>-NMF and FCC Algorithms in Long-Wave Hyperspectral Infrared ImageryBardia Yousefi0Clemente Ibarra-Castanedo1Martin Chamberland2Xavier P. V. Maldague3Georges Beaudoin4Computer Vision and System Laboratory (CVSL), Department of Electrical and Computer Engineering, Laval University, Québec, QC G2E 6J5, CanadaComputer Vision and System Laboratory (CVSL), Department of Electrical and Computer Engineering, Laval University, Québec, QC G2E 6J5, CanadaTelops Inc., 100-2600 St-Jean-Baptiste Ave, Québec, QC G2E 6J5, CanadaComputer Vision and System Laboratory (CVSL), Department of Electrical and Computer Engineering, Laval University, Québec, QC G2E 6J5, CanadaDepartment of Geology and Geological Engineering, Laval University, Québec, QC G2E 6J5, CanadaClustering methods unequivocally show considerable influence on many recent algorithms and play an important role in hyperspectral data analysis. Here, we challenge the clustering for mineral identification using two different strategies in hyperspectral long wave infrared (LWIR, 7.7–11.8 μm). For that, we compare two algorithms to perform the mineral identification in a unique dataset. The first algorithm uses spectral comparison techniques for all the pixel-spectra and creates RGB false color composites (FCC). Then, a color based clustering is used to group the regions (called FCC-clustering). The second algorithm clusters all the pixel-spectra to directly group the spectra. Then, the first rank of non-negative matrix factorization (NMF) extracts the representative of each cluster and compares results with the spectral library of JPL/NASA. These techniques give the comparison values as features which convert into RGB-FCC as the results (called clustering rank<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>1</mn></msup></semantics></math></inline-formula>-NMF). We applied K-means as clustering approach, which can be modified in any other similar clustering approach. The results of the clustering-rank<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>1</mn></msup></semantics></math></inline-formula>-NMF algorithm indicate significant computational efficiency (more than 20 times faster than the previous approach) and promising performance for mineral identification having up to 75.8% and 84.8% average accuracies for FCC-clustering and clustering-rank<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>1</mn></msup></semantics></math></inline-formula> NMF algorithms (using spectral angle mapper (SAM)), respectively. Furthermore, several spectral comparison techniques are used also such as adaptive matched subspace detector (AMSD), orthogonal subspace projection (OSP) algorithm, principal component analysis (PCA), local matched filter (PLMF), SAM, and normalized cross correlation (NCC) for both algorithms and most of them show a similar range in accuracy. However, SAM and NCC are preferred due to their computational simplicity. Our algorithms strive to identify eleven different mineral grains (biotite, diopside, epidote, goethite, kyanite, scheelite, smithsonite, tourmaline, pyrope, olivine, and quartz).https://www.mdpi.com/2072-4292/13/11/2125long-wave infrared hyperspectral imagingmineral identificationclustering of hyperspectral dataspectral comparison method |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Bardia Yousefi Clemente Ibarra-Castanedo Martin Chamberland Xavier P. V. Maldague Georges Beaudoin |
spellingShingle |
Bardia Yousefi Clemente Ibarra-Castanedo Martin Chamberland Xavier P. V. Maldague Georges Beaudoin Unsupervised Identification of Targeted Spectra Applying Rank<sup>1</sup>-NMF and FCC Algorithms in Long-Wave Hyperspectral Infrared Imagery Remote Sensing long-wave infrared hyperspectral imaging mineral identification clustering of hyperspectral data spectral comparison method |
author_facet |
Bardia Yousefi Clemente Ibarra-Castanedo Martin Chamberland Xavier P. V. Maldague Georges Beaudoin |
author_sort |
Bardia Yousefi |
title |
Unsupervised Identification of Targeted Spectra Applying Rank<sup>1</sup>-NMF and FCC Algorithms in Long-Wave Hyperspectral Infrared Imagery |
title_short |
Unsupervised Identification of Targeted Spectra Applying Rank<sup>1</sup>-NMF and FCC Algorithms in Long-Wave Hyperspectral Infrared Imagery |
title_full |
Unsupervised Identification of Targeted Spectra Applying Rank<sup>1</sup>-NMF and FCC Algorithms in Long-Wave Hyperspectral Infrared Imagery |
title_fullStr |
Unsupervised Identification of Targeted Spectra Applying Rank<sup>1</sup>-NMF and FCC Algorithms in Long-Wave Hyperspectral Infrared Imagery |
title_full_unstemmed |
Unsupervised Identification of Targeted Spectra Applying Rank<sup>1</sup>-NMF and FCC Algorithms in Long-Wave Hyperspectral Infrared Imagery |
title_sort |
unsupervised identification of targeted spectra applying rank<sup>1</sup>-nmf and fcc algorithms in long-wave hyperspectral infrared imagery |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2021-05-01 |
description |
Clustering methods unequivocally show considerable influence on many recent algorithms and play an important role in hyperspectral data analysis. Here, we challenge the clustering for mineral identification using two different strategies in hyperspectral long wave infrared (LWIR, 7.7–11.8 μm). For that, we compare two algorithms to perform the mineral identification in a unique dataset. The first algorithm uses spectral comparison techniques for all the pixel-spectra and creates RGB false color composites (FCC). Then, a color based clustering is used to group the regions (called FCC-clustering). The second algorithm clusters all the pixel-spectra to directly group the spectra. Then, the first rank of non-negative matrix factorization (NMF) extracts the representative of each cluster and compares results with the spectral library of JPL/NASA. These techniques give the comparison values as features which convert into RGB-FCC as the results (called clustering rank<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>1</mn></msup></semantics></math></inline-formula>-NMF). We applied K-means as clustering approach, which can be modified in any other similar clustering approach. The results of the clustering-rank<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>1</mn></msup></semantics></math></inline-formula>-NMF algorithm indicate significant computational efficiency (more than 20 times faster than the previous approach) and promising performance for mineral identification having up to 75.8% and 84.8% average accuracies for FCC-clustering and clustering-rank<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>1</mn></msup></semantics></math></inline-formula> NMF algorithms (using spectral angle mapper (SAM)), respectively. Furthermore, several spectral comparison techniques are used also such as adaptive matched subspace detector (AMSD), orthogonal subspace projection (OSP) algorithm, principal component analysis (PCA), local matched filter (PLMF), SAM, and normalized cross correlation (NCC) for both algorithms and most of them show a similar range in accuracy. However, SAM and NCC are preferred due to their computational simplicity. Our algorithms strive to identify eleven different mineral grains (biotite, diopside, epidote, goethite, kyanite, scheelite, smithsonite, tourmaline, pyrope, olivine, and quartz). |
topic |
long-wave infrared hyperspectral imaging mineral identification clustering of hyperspectral data spectral comparison method |
url |
https://www.mdpi.com/2072-4292/13/11/2125 |
work_keys_str_mv |
AT bardiayousefi unsupervisedidentificationoftargetedspectraapplyingranksup1supnmfandfccalgorithmsinlongwavehyperspectralinfraredimagery AT clementeibarracastanedo unsupervisedidentificationoftargetedspectraapplyingranksup1supnmfandfccalgorithmsinlongwavehyperspectralinfraredimagery AT martinchamberland unsupervisedidentificationoftargetedspectraapplyingranksup1supnmfandfccalgorithmsinlongwavehyperspectralinfraredimagery AT xavierpvmaldague unsupervisedidentificationoftargetedspectraapplyingranksup1supnmfandfccalgorithmsinlongwavehyperspectralinfraredimagery AT georgesbeaudoin unsupervisedidentificationoftargetedspectraapplyingranksup1supnmfandfccalgorithmsinlongwavehyperspectralinfraredimagery |
_version_ |
1721412241360683008 |