Spectral variability within species and its effects on savanna tree species discrimination

Moses A. Cho, Pravesh Debba, Renaud Mathieu, Jan Van Aardt, Greg Asner, Laven Naidoo, Russell Main, Abel Ramoelo, Bongani Majeke

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Differences in within-species phenology and structure driven by factors including topography, edaphic properties, and climatic variables present important challenges for species differentiation with remote sensing in the Kruger National Park, South Africa. The objective of this study was to examine probable factors including intraspecies spectral variability and the spectral sample size that could affect remote sensing of Savanna tree species across a land-use gradient in the Kruger National park. Eighteen species were examined: Acacia gerradii, Acacia nigrescens, Combretum apiculatum, Combretum collinum, Combretum hereroense, Combretum imberbe, Combretum zeyheri, Dichrostachys cinerea, Euclea sp (E. divinurum and E. natalensis, Gymnosporia sp (G. buxifolia and G. senegalensis), Lonchocarpus capassa, Peltoforum africanum, Piliostigma thonningii, Pterocarpus rotundifolia, Sclerocarya birrea, Strychnos sp (S. madagascariensis, S. usambarensis), Terminalia sericea and Ziziphus mucronata. Discriminating species using the K-nearest neighbour (K = 1) classifier with spectral angle mapper (SAM) yielded a higher classification accuracy (48% overall accuracy) compared to 16% for the classification involving the mean spectra for each species as the training spectral set. Within-species spectral variability and the training sample size were identified as important factors affecting classification accuracy of the tree species. We recommend a non-parametric classifier such as K-nearest neighbour classifier for classifying and mapping tree species in a highly complex environment such as the savanna system of the Kruger National Park.

Original languageEnglish (US)
Title of host publication2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Proceedings
PagesII190-II193
DOIs
StatePublished - Dec 1 2009
Externally publishedYes
Event2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Cape Town, South Africa
Duration: Jul 12 2009Jul 17 2009

Publication series

NameInternational Geoscience and Remote Sensing Symposium (IGARSS)
Volume2

Other

Other2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009
CountrySouth Africa
CityCape Town
Period7/12/097/17/09

Fingerprint

savanna
Classifiers
Remote sensing
Land use
Topography
national park
remote sensing
effect
phenology
topography
land use

Keywords

  • Multiple endmember approach
  • Savanna tree species
  • Spectral angle mapper
  • Spectral variability

ASJC Scopus subject areas

  • Computer Science Applications
  • Earth and Planetary Sciences(all)

Cite this

Cho, M. A., Debba, P., Mathieu, R., Van Aardt, J., Asner, G., Naidoo, L., ... Majeke, B. (2009). Spectral variability within species and its effects on savanna tree species discrimination. In 2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Proceedings (pp. II190-II193). [5418038] (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2). https://doi.org/10.1109/IGARSS.2009.5418038

Spectral variability within species and its effects on savanna tree species discrimination. / Cho, Moses A.; Debba, Pravesh; Mathieu, Renaud; Van Aardt, Jan; Asner, Greg; Naidoo, Laven; Main, Russell; Ramoelo, Abel; Majeke, Bongani.

2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Proceedings. 2009. p. II190-II193 5418038 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Cho, MA, Debba, P, Mathieu, R, Van Aardt, J, Asner, G, Naidoo, L, Main, R, Ramoelo, A & Majeke, B 2009, Spectral variability within species and its effects on savanna tree species discrimination. in 2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Proceedings., 5418038, International Geoscience and Remote Sensing Symposium (IGARSS), vol. 2, pp. II190-II193, 2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009, Cape Town, South Africa, 7/12/09. https://doi.org/10.1109/IGARSS.2009.5418038
Cho MA, Debba P, Mathieu R, Van Aardt J, Asner G, Naidoo L et al. Spectral variability within species and its effects on savanna tree species discrimination. In 2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Proceedings. 2009. p. II190-II193. 5418038. (International Geoscience and Remote Sensing Symposium (IGARSS)). https://doi.org/10.1109/IGARSS.2009.5418038
Cho, Moses A. ; Debba, Pravesh ; Mathieu, Renaud ; Van Aardt, Jan ; Asner, Greg ; Naidoo, Laven ; Main, Russell ; Ramoelo, Abel ; Majeke, Bongani. / Spectral variability within species and its effects on savanna tree species discrimination. 2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Proceedings. 2009. pp. II190-II193 (International Geoscience and Remote Sensing Symposium (IGARSS)).
@inproceedings{6655a216c29d48f59564c5366afd42cc,
title = "Spectral variability within species and its effects on savanna tree species discrimination",
abstract = "Differences in within-species phenology and structure driven by factors including topography, edaphic properties, and climatic variables present important challenges for species differentiation with remote sensing in the Kruger National Park, South Africa. The objective of this study was to examine probable factors including intraspecies spectral variability and the spectral sample size that could affect remote sensing of Savanna tree species across a land-use gradient in the Kruger National park. Eighteen species were examined: Acacia gerradii, Acacia nigrescens, Combretum apiculatum, Combretum collinum, Combretum hereroense, Combretum imberbe, Combretum zeyheri, Dichrostachys cinerea, Euclea sp (E. divinurum and E. natalensis, Gymnosporia sp (G. buxifolia and G. senegalensis), Lonchocarpus capassa, Peltoforum africanum, Piliostigma thonningii, Pterocarpus rotundifolia, Sclerocarya birrea, Strychnos sp (S. madagascariensis, S. usambarensis), Terminalia sericea and Ziziphus mucronata. Discriminating species using the K-nearest neighbour (K = 1) classifier with spectral angle mapper (SAM) yielded a higher classification accuracy (48{\%} overall accuracy) compared to 16{\%} for the classification involving the mean spectra for each species as the training spectral set. Within-species spectral variability and the training sample size were identified as important factors affecting classification accuracy of the tree species. We recommend a non-parametric classifier such as K-nearest neighbour classifier for classifying and mapping tree species in a highly complex environment such as the savanna system of the Kruger National Park.",
keywords = "Multiple endmember approach, Savanna tree species, Spectral angle mapper, Spectral variability",
author = "Cho, {Moses A.} and Pravesh Debba and Renaud Mathieu and {Van Aardt}, Jan and Greg Asner and Laven Naidoo and Russell Main and Abel Ramoelo and Bongani Majeke",
year = "2009",
month = "12",
day = "1",
doi = "10.1109/IGARSS.2009.5418038",
language = "English (US)",
isbn = "9781424433957",
series = "International Geoscience and Remote Sensing Symposium (IGARSS)",
pages = "II190--II193",
booktitle = "2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Proceedings",

}

TY - GEN

T1 - Spectral variability within species and its effects on savanna tree species discrimination

AU - Cho, Moses A.

AU - Debba, Pravesh

AU - Mathieu, Renaud

AU - Van Aardt, Jan

AU - Asner, Greg

AU - Naidoo, Laven

AU - Main, Russell

AU - Ramoelo, Abel

AU - Majeke, Bongani

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Differences in within-species phenology and structure driven by factors including topography, edaphic properties, and climatic variables present important challenges for species differentiation with remote sensing in the Kruger National Park, South Africa. The objective of this study was to examine probable factors including intraspecies spectral variability and the spectral sample size that could affect remote sensing of Savanna tree species across a land-use gradient in the Kruger National park. Eighteen species were examined: Acacia gerradii, Acacia nigrescens, Combretum apiculatum, Combretum collinum, Combretum hereroense, Combretum imberbe, Combretum zeyheri, Dichrostachys cinerea, Euclea sp (E. divinurum and E. natalensis, Gymnosporia sp (G. buxifolia and G. senegalensis), Lonchocarpus capassa, Peltoforum africanum, Piliostigma thonningii, Pterocarpus rotundifolia, Sclerocarya birrea, Strychnos sp (S. madagascariensis, S. usambarensis), Terminalia sericea and Ziziphus mucronata. Discriminating species using the K-nearest neighbour (K = 1) classifier with spectral angle mapper (SAM) yielded a higher classification accuracy (48% overall accuracy) compared to 16% for the classification involving the mean spectra for each species as the training spectral set. Within-species spectral variability and the training sample size were identified as important factors affecting classification accuracy of the tree species. We recommend a non-parametric classifier such as K-nearest neighbour classifier for classifying and mapping tree species in a highly complex environment such as the savanna system of the Kruger National Park.

AB - Differences in within-species phenology and structure driven by factors including topography, edaphic properties, and climatic variables present important challenges for species differentiation with remote sensing in the Kruger National Park, South Africa. The objective of this study was to examine probable factors including intraspecies spectral variability and the spectral sample size that could affect remote sensing of Savanna tree species across a land-use gradient in the Kruger National park. Eighteen species were examined: Acacia gerradii, Acacia nigrescens, Combretum apiculatum, Combretum collinum, Combretum hereroense, Combretum imberbe, Combretum zeyheri, Dichrostachys cinerea, Euclea sp (E. divinurum and E. natalensis, Gymnosporia sp (G. buxifolia and G. senegalensis), Lonchocarpus capassa, Peltoforum africanum, Piliostigma thonningii, Pterocarpus rotundifolia, Sclerocarya birrea, Strychnos sp (S. madagascariensis, S. usambarensis), Terminalia sericea and Ziziphus mucronata. Discriminating species using the K-nearest neighbour (K = 1) classifier with spectral angle mapper (SAM) yielded a higher classification accuracy (48% overall accuracy) compared to 16% for the classification involving the mean spectra for each species as the training spectral set. Within-species spectral variability and the training sample size were identified as important factors affecting classification accuracy of the tree species. We recommend a non-parametric classifier such as K-nearest neighbour classifier for classifying and mapping tree species in a highly complex environment such as the savanna system of the Kruger National Park.

KW - Multiple endmember approach

KW - Savanna tree species

KW - Spectral angle mapper

KW - Spectral variability

UR - http://www.scopus.com/inward/record.url?scp=77951132574&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951132574&partnerID=8YFLogxK

U2 - 10.1109/IGARSS.2009.5418038

DO - 10.1109/IGARSS.2009.5418038

M3 - Conference contribution

AN - SCOPUS:77951132574

SN - 9781424433957

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - II190-II193

BT - 2009 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2009 - Proceedings

ER -