Spectral Complexity of Hyperspectral Images: A New Approach for Mangrove Classification
Hyperspectral remote sensing across multiple spatio-temporal scales allows for mapping and monitoring mangrove habitats to support urgent conservation efforts. The use of hyperspectral imagery for assessing mangroves is less common than for terrestrial forest ecosystems. In this study, two well-know...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-07-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/13/13/2604 |
id |
doaj-83b17f513eb74b6382ba0ad859e1a4e4 |
---|---|
record_format |
Article |
spelling |
doaj-83b17f513eb74b6382ba0ad859e1a4e42021-07-15T15:44:39ZengMDPI AGRemote Sensing2072-42922021-07-01132604260410.3390/rs13132604Spectral Complexity of Hyperspectral Images: A New Approach for Mangrove ClassificationPatrick Osei Darko0Margaret Kalacska1J. Pablo Arroyo-Mora2Matthew E. Fagan3Applied Remote Sensing Laboratory, Department of Geography, McGill University, Montréal, QC H3A 0B9, CanadaApplied Remote Sensing Laboratory, Department of Geography, McGill University, Montréal, QC H3A 0B9, CanadaFlight Research Laboratory, National Research Council of Canada, Ottawa, ON K1A 0R6, CanadaDepartment of Geography and Environmental Systems, University of Maryland, Baltimore County (UMBC), Baltimore, MD 21250, USAHyperspectral remote sensing across multiple spatio-temporal scales allows for mapping and monitoring mangrove habitats to support urgent conservation efforts. The use of hyperspectral imagery for assessing mangroves is less common than for terrestrial forest ecosystems. In this study, two well-known measures in statistical physics, Mean Information Gain (MIG) and Marginal Entropy (ME), have been adapted to high spatial resolution (2.5 m) full range (Visible-Shortwave-Infrared) airborne hyperspectral imagery. These two spectral complexity metrics describe the spatial heterogeneity and the aspatial heterogeneity of the reflectance. In this study, we compare MIG and ME with surface reflectance for mapping mangrove extent and species composition in the Sierpe mangroves in Costa Rica. The highest accuracy for separating mangroves from forest was achieved with visible-near infrared (VNIR) reflectance (98.8% overall accuracy), following by shortwave infrared (SWIR) MIG and ME (98%). Our results also show that MIG and ME can discriminate dominant mangrove species with higher accuracy than surface reflectance alone (e.g., MIG–VNIR = 93.6% vs. VNIR Reflectance = 89.7%).https://www.mdpi.com/2072-4292/13/13/2604aspatial heterogeneityspatial heterogeneityspecies discriminationairbornemean information gainmarginal entropy |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Patrick Osei Darko Margaret Kalacska J. Pablo Arroyo-Mora Matthew E. Fagan |
spellingShingle |
Patrick Osei Darko Margaret Kalacska J. Pablo Arroyo-Mora Matthew E. Fagan Spectral Complexity of Hyperspectral Images: A New Approach for Mangrove Classification Remote Sensing aspatial heterogeneity spatial heterogeneity species discrimination airborne mean information gain marginal entropy |
author_facet |
Patrick Osei Darko Margaret Kalacska J. Pablo Arroyo-Mora Matthew E. Fagan |
author_sort |
Patrick Osei Darko |
title |
Spectral Complexity of Hyperspectral Images: A New Approach for Mangrove Classification |
title_short |
Spectral Complexity of Hyperspectral Images: A New Approach for Mangrove Classification |
title_full |
Spectral Complexity of Hyperspectral Images: A New Approach for Mangrove Classification |
title_fullStr |
Spectral Complexity of Hyperspectral Images: A New Approach for Mangrove Classification |
title_full_unstemmed |
Spectral Complexity of Hyperspectral Images: A New Approach for Mangrove Classification |
title_sort |
spectral complexity of hyperspectral images: a new approach for mangrove classification |
publisher |
MDPI AG |
series |
Remote Sensing |
issn |
2072-4292 |
publishDate |
2021-07-01 |
description |
Hyperspectral remote sensing across multiple spatio-temporal scales allows for mapping and monitoring mangrove habitats to support urgent conservation efforts. The use of hyperspectral imagery for assessing mangroves is less common than for terrestrial forest ecosystems. In this study, two well-known measures in statistical physics, Mean Information Gain (MIG) and Marginal Entropy (ME), have been adapted to high spatial resolution (2.5 m) full range (Visible-Shortwave-Infrared) airborne hyperspectral imagery. These two spectral complexity metrics describe the spatial heterogeneity and the aspatial heterogeneity of the reflectance. In this study, we compare MIG and ME with surface reflectance for mapping mangrove extent and species composition in the Sierpe mangroves in Costa Rica. The highest accuracy for separating mangroves from forest was achieved with visible-near infrared (VNIR) reflectance (98.8% overall accuracy), following by shortwave infrared (SWIR) MIG and ME (98%). Our results also show that MIG and ME can discriminate dominant mangrove species with higher accuracy than surface reflectance alone (e.g., MIG–VNIR = 93.6% vs. VNIR Reflectance = 89.7%). |
topic |
aspatial heterogeneity spatial heterogeneity species discrimination airborne mean information gain marginal entropy |
url |
https://www.mdpi.com/2072-4292/13/13/2604 |
work_keys_str_mv |
AT patrickoseidarko spectralcomplexityofhyperspectralimagesanewapproachformangroveclassification AT margaretkalacska spectralcomplexityofhyperspectralimagesanewapproachformangroveclassification AT jpabloarroyomora spectralcomplexityofhyperspectralimagesanewapproachformangroveclassification AT matthewefagan spectralcomplexityofhyperspectralimagesanewapproachformangroveclassification |
_version_ |
1721298532646780928 |