Classification of contamination in salt marsh plants using hyperspectral reflectance

Machelle D. Wilson, Susan L. Ustin, David M Rocke

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

In this paper, we compare the classification effectiveness of two relatively new techniques on data consisting of leaf-level reflectance from five species of salt marsh and two species of crop plants (in four experiments) that have been exposed to varying levels of different heavy metal or petroleum toxicity, with a control treatment for each experiment. If these methodologies work well on leaf-level data, then there is hope that they will also work well on data from air- and spaceborne platforms. The classification methods compared were support vector classification (SVC) of exposed and nonexposed plants based on the spectral reflectance data, and partial least squares compression of the spectral reflectance data followed by classification using logistic discrimination (PLS/LD). The statistic we used to compare the effectiveness of the methodologies was the leave-one-out cross-validation estimate of the prediction error. Our results suggest that both techniques perform reasonably well, but that SVC was superior to PLS/LD for use on hyperspectral data and it is worth exploring as a technique for classifying heavy-metal or petroleum exposed plants for the more complicated data from air- and spaceborne sensors.

Original languageEnglish (US)
Pages (from-to)1088-1095
Number of pages8
JournalIEEE Transactions on Geoscience and Remote Sensing
Volume42
Issue number5
DOIs
StatePublished - May 2004

Fingerprint

marshlands
saltmarsh
reflectance
contamination
Contamination
Salts
salts
spectral reflectance
Petroleum
heavy metals
Heavy Metals
crude oil
leaves
Heavy metals
Crude oil
methodology
air
crops
logistics
petroleum

Keywords

  • Heavy metals
  • Hyperspectral
  • Logistic discrimination (LD)
  • Partial least squares (PLS)
  • Petroleum
  • Reflectance
  • Remote sensing
  • Support vector machines (SVMs)

ASJC Scopus subject areas

  • Geochemistry and Petrology
  • Geophysics
  • Computers in Earth Sciences
  • Electrical and Electronic Engineering

Cite this

Classification of contamination in salt marsh plants using hyperspectral reflectance. / Wilson, Machelle D.; Ustin, Susan L.; Rocke, David M.

In: IEEE Transactions on Geoscience and Remote Sensing, Vol. 42, No. 5, 05.2004, p. 1088-1095.

Research output: Contribution to journalArticle

@article{a46cda87b5964ba699f6e56a95683647,
title = "Classification of contamination in salt marsh plants using hyperspectral reflectance",
abstract = "In this paper, we compare the classification effectiveness of two relatively new techniques on data consisting of leaf-level reflectance from five species of salt marsh and two species of crop plants (in four experiments) that have been exposed to varying levels of different heavy metal or petroleum toxicity, with a control treatment for each experiment. If these methodologies work well on leaf-level data, then there is hope that they will also work well on data from air- and spaceborne platforms. The classification methods compared were support vector classification (SVC) of exposed and nonexposed plants based on the spectral reflectance data, and partial least squares compression of the spectral reflectance data followed by classification using logistic discrimination (PLS/LD). The statistic we used to compare the effectiveness of the methodologies was the leave-one-out cross-validation estimate of the prediction error. Our results suggest that both techniques perform reasonably well, but that SVC was superior to PLS/LD for use on hyperspectral data and it is worth exploring as a technique for classifying heavy-metal or petroleum exposed plants for the more complicated data from air- and spaceborne sensors.",
keywords = "Heavy metals, Hyperspectral, Logistic discrimination (LD), Partial least squares (PLS), Petroleum, Reflectance, Remote sensing, Support vector machines (SVMs)",
author = "Wilson, {Machelle D.} and Ustin, {Susan L.} and Rocke, {David M}",
year = "2004",
month = "5",
doi = "10.1109/TGRS.2003.823278",
language = "English (US)",
volume = "42",
pages = "1088--1095",
journal = "IEEE Transactions on Geoscience and Remote Sensing",
issn = "0196-2892",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Classification of contamination in salt marsh plants using hyperspectral reflectance

AU - Wilson, Machelle D.

AU - Ustin, Susan L.

AU - Rocke, David M

PY - 2004/5

Y1 - 2004/5

N2 - In this paper, we compare the classification effectiveness of two relatively new techniques on data consisting of leaf-level reflectance from five species of salt marsh and two species of crop plants (in four experiments) that have been exposed to varying levels of different heavy metal or petroleum toxicity, with a control treatment for each experiment. If these methodologies work well on leaf-level data, then there is hope that they will also work well on data from air- and spaceborne platforms. The classification methods compared were support vector classification (SVC) of exposed and nonexposed plants based on the spectral reflectance data, and partial least squares compression of the spectral reflectance data followed by classification using logistic discrimination (PLS/LD). The statistic we used to compare the effectiveness of the methodologies was the leave-one-out cross-validation estimate of the prediction error. Our results suggest that both techniques perform reasonably well, but that SVC was superior to PLS/LD for use on hyperspectral data and it is worth exploring as a technique for classifying heavy-metal or petroleum exposed plants for the more complicated data from air- and spaceborne sensors.

AB - In this paper, we compare the classification effectiveness of two relatively new techniques on data consisting of leaf-level reflectance from five species of salt marsh and two species of crop plants (in four experiments) that have been exposed to varying levels of different heavy metal or petroleum toxicity, with a control treatment for each experiment. If these methodologies work well on leaf-level data, then there is hope that they will also work well on data from air- and spaceborne platforms. The classification methods compared were support vector classification (SVC) of exposed and nonexposed plants based on the spectral reflectance data, and partial least squares compression of the spectral reflectance data followed by classification using logistic discrimination (PLS/LD). The statistic we used to compare the effectiveness of the methodologies was the leave-one-out cross-validation estimate of the prediction error. Our results suggest that both techniques perform reasonably well, but that SVC was superior to PLS/LD for use on hyperspectral data and it is worth exploring as a technique for classifying heavy-metal or petroleum exposed plants for the more complicated data from air- and spaceborne sensors.

KW - Heavy metals

KW - Hyperspectral

KW - Logistic discrimination (LD)

KW - Partial least squares (PLS)

KW - Petroleum

KW - Reflectance

KW - Remote sensing

KW - Support vector machines (SVMs)

UR - http://www.scopus.com/inward/record.url?scp=2642563585&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2642563585&partnerID=8YFLogxK

U2 - 10.1109/TGRS.2003.823278

DO - 10.1109/TGRS.2003.823278

M3 - Article

VL - 42

SP - 1088

EP - 1095

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

SN - 0196-2892

IS - 5

ER -