Tree species discrimination in tropical forests using airborne imaging spectroscopy

Jean Baptiste Feret, Gregory P. Asner

Research output: Contribution to journalArticle

121 Citations (Scopus)

Abstract

We identify canopy species in a Hawaiian tropical forest using supervised classification applied to airborne hyperspectral imagery acquired with the Carnegie Airborne Observatory-Alpha system. Nonparametric methods (linear and radial basis function support vector machine, artificial neural network, and k-nearest neighbor) and parametric methods (linear, quadratic, and regularized discriminant analysis) are compared for a range of species richness values and training sample sizes. We find a clear advantage in using regularized discriminant analysis, linear discriminant analysis, and support vector machines. No unique optimal classifier was found for all conditions tested, but we highlight the possibility of improving support vector machine classification with a better optimization of its free parameters. We also confirm that a combination of spectral and spatial information increases accuracy of species classification: we combine segmentation and species classification from regularized discriminant analysis to produce a map of the 17 discriminated species. Finally, we compare different methods to assess spectral separability and find a better ability of Bhattacharyya distance to assess separability within and among species. The results indicate that species mapping is tractable in tropical forests when using high-fidelity imaging spectroscopy.

Original languageEnglish (US)
Article number6241414
Pages (from-to)73-84
Number of pages12
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume51
Issue number1
DOIs
StatePublished - Jan 1 2013
Externally publishedYes

Fingerprint

Discriminant analysis
tropical forest
spectroscopy
Spectroscopy
discriminant analysis
Support vector machines
Imaging techniques
Observatories
airborne sensing
Classifiers
image classification
Neural networks
artificial neural network
segmentation
observatory
species richness
canopy
method
support vector machine

Keywords

  • Carnegie Airborne Observatory (CAO)
  • hyperspectral imaging
  • image classification
  • tree species identification
  • tropical biodiversity

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Earth and Planetary Sciences(all)

Cite this

Tree species discrimination in tropical forests using airborne imaging spectroscopy. / Feret, Jean Baptiste; Asner, Gregory P.

In: IEEE Transactions on Geoscience and Remote Sensing, Vol. 51, No. 1, 6241414, 01.01.2013, p. 73-84.

Research output: Contribution to journalArticle

@article{73e32d118b9e4eec95384881d646caa8,
title = "Tree species discrimination in tropical forests using airborne imaging spectroscopy",
abstract = "We identify canopy species in a Hawaiian tropical forest using supervised classification applied to airborne hyperspectral imagery acquired with the Carnegie Airborne Observatory-Alpha system. Nonparametric methods (linear and radial basis function support vector machine, artificial neural network, and k-nearest neighbor) and parametric methods (linear, quadratic, and regularized discriminant analysis) are compared for a range of species richness values and training sample sizes. We find a clear advantage in using regularized discriminant analysis, linear discriminant analysis, and support vector machines. No unique optimal classifier was found for all conditions tested, but we highlight the possibility of improving support vector machine classification with a better optimization of its free parameters. We also confirm that a combination of spectral and spatial information increases accuracy of species classification: we combine segmentation and species classification from regularized discriminant analysis to produce a map of the 17 discriminated species. Finally, we compare different methods to assess spectral separability and find a better ability of Bhattacharyya distance to assess separability within and among species. The results indicate that species mapping is tractable in tropical forests when using high-fidelity imaging spectroscopy.",
keywords = "Carnegie Airborne Observatory (CAO), hyperspectral imaging, image classification, tree species identification, tropical biodiversity",
author = "Feret, {Jean Baptiste} and Asner, {Gregory P.}",
year = "2013",
month = "1",
day = "1",
doi = "10.1109/TGRS.2012.2199323",
language = "English (US)",
volume = "51",
pages = "73--84",
journal = "IEEE Transactions on Geoscience and Remote Sensing",
issn = "0196-2892",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Tree species discrimination in tropical forests using airborne imaging spectroscopy

AU - Feret, Jean Baptiste

AU - Asner, Gregory P.

PY - 2013/1/1

Y1 - 2013/1/1

N2 - We identify canopy species in a Hawaiian tropical forest using supervised classification applied to airborne hyperspectral imagery acquired with the Carnegie Airborne Observatory-Alpha system. Nonparametric methods (linear and radial basis function support vector machine, artificial neural network, and k-nearest neighbor) and parametric methods (linear, quadratic, and regularized discriminant analysis) are compared for a range of species richness values and training sample sizes. We find a clear advantage in using regularized discriminant analysis, linear discriminant analysis, and support vector machines. No unique optimal classifier was found for all conditions tested, but we highlight the possibility of improving support vector machine classification with a better optimization of its free parameters. We also confirm that a combination of spectral and spatial information increases accuracy of species classification: we combine segmentation and species classification from regularized discriminant analysis to produce a map of the 17 discriminated species. Finally, we compare different methods to assess spectral separability and find a better ability of Bhattacharyya distance to assess separability within and among species. The results indicate that species mapping is tractable in tropical forests when using high-fidelity imaging spectroscopy.

AB - We identify canopy species in a Hawaiian tropical forest using supervised classification applied to airborne hyperspectral imagery acquired with the Carnegie Airborne Observatory-Alpha system. Nonparametric methods (linear and radial basis function support vector machine, artificial neural network, and k-nearest neighbor) and parametric methods (linear, quadratic, and regularized discriminant analysis) are compared for a range of species richness values and training sample sizes. We find a clear advantage in using regularized discriminant analysis, linear discriminant analysis, and support vector machines. No unique optimal classifier was found for all conditions tested, but we highlight the possibility of improving support vector machine classification with a better optimization of its free parameters. We also confirm that a combination of spectral and spatial information increases accuracy of species classification: we combine segmentation and species classification from regularized discriminant analysis to produce a map of the 17 discriminated species. Finally, we compare different methods to assess spectral separability and find a better ability of Bhattacharyya distance to assess separability within and among species. The results indicate that species mapping is tractable in tropical forests when using high-fidelity imaging spectroscopy.

KW - Carnegie Airborne Observatory (CAO)

KW - hyperspectral imaging

KW - image classification

KW - tree species identification

KW - tropical biodiversity

UR - http://www.scopus.com/inward/record.url?scp=84871745702&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871745702&partnerID=8YFLogxK

U2 - 10.1109/TGRS.2012.2199323

DO - 10.1109/TGRS.2012.2199323

M3 - Article

AN - SCOPUS:84871745702

VL - 51

SP - 73

EP - 84

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

SN - 0196-2892

IS - 1

M1 - 6241414

ER -