Multiple classifiers applied to multisource remote sensing data

Gunnar Jakob Briem; Jon Atli Benediktsson; Johannes R. Sveinsson

doi:10.1109/TGRS.2002.802476

Multiple classifiers applied to multisource remote sensing data

Gunnar Jakob Briem^*, Jon Atli Benediktsson, Johannes R. Sveinsson

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

224 Citations (Scopus)

Abstract

The combination of multisource remote sensing and geographic data is believed to offer improved accuracies in land cover classification. For such classification, the conventional parametric statistical classifiers, which have been applied successfully in remote sensing for the last two decades, are not appropriate, since a convenient multivariate statistical model does not exist for the data. In this paper, several single and multiple classifiers, that are appropriate for the classification of multisource remote sensing and geographic data are considered. The focus is on multiple classifiers: bagging algorithms, boosting algorithms, and consensus-theoretic classifiers. These multiple classifiers have different characteristics. The performance of the algorithms in terms of accuracies is compared for two multisource remote sensing and geographic datasets. In the experiments, the multiple classifiers outperform the single classifiers in terms of overall accuracies.

Original language	English
Pages (from-to)	2291-2299
Number of pages	9
Journal	IEEE Transactions on Geoscience and Remote Sensing
Volume	40
Issue number	10
DOIs	https://doi.org/10.1109/TGRS.2002.802476
Publication status	Published - Oct 2002

Bibliographical note

Funding Information:
Manuscript received September 26, 2001; revised May 17, 2002. This research was supported in part by the Icelandic Research Council and the Research Fund of the University of Iceland. G. J. Briem was with the Department of Electrical and Computer Engineering, University of Iceland, Reykjavik IS-107, Iceland. He is now with Dimon Software, Reykjavik IS-105, Iceland. J. A. Benediktsson and J. R. Sveinsson are with the Department of Electrical and Computer Engineering, University of Iceland, Reykjavik IS-107, Iceland. Digital Object Identifier 10.1109/TGRS.2002.802476

Other keywords

Bagging
Boosting
Consensus theory
Multiple classifiers
Multisource remote sensing data

Access to Document

10.1109/TGRS.2002.802476

Cite this

@article{cf8af5540d7a49e880a3e069c52403d4,

title = "Multiple classifiers applied to multisource remote sensing data",

abstract = "The combination of multisource remote sensing and geographic data is believed to offer improved accuracies in land cover classification. For such classification, the conventional parametric statistical classifiers, which have been applied successfully in remote sensing for the last two decades, are not appropriate, since a convenient multivariate statistical model does not exist for the data. In this paper, several single and multiple classifiers, that are appropriate for the classification of multisource remote sensing and geographic data are considered. The focus is on multiple classifiers: bagging algorithms, boosting algorithms, and consensus-theoretic classifiers. These multiple classifiers have different characteristics. The performance of the algorithms in terms of accuracies is compared for two multisource remote sensing and geographic datasets. In the experiments, the multiple classifiers outperform the single classifiers in terms of overall accuracies.",

keywords = "Bagging, Boosting, Consensus theory, Multiple classifiers, Multisource remote sensing data",

author = "Briem, {Gunnar Jakob} and Benediktsson, {Jon Atli} and Sveinsson, {Johannes R.}",

note = "Funding Information: Manuscript received September 26, 2001; revised May 17, 2002. This research was supported in part by the Icelandic Research Council and the Research Fund of the University of Iceland. G. J. Briem was with the Department of Electrical and Computer Engineering, University of Iceland, Reykjavik IS-107, Iceland. He is now with Dimon Software, Reykjavik IS-105, Iceland. J. A. Benediktsson and J. R. Sveinsson are with the Department of Electrical and Computer Engineering, University of Iceland, Reykjavik IS-107, Iceland. Digital Object Identifier 10.1109/TGRS.2002.802476",

year = "2002",

month = oct,

doi = "10.1109/TGRS.2002.802476",

language = "English",

volume = "40",

pages = "2291--2299",

journal = "IEEE Transactions on Geoscience and Remote Sensing",

issn = "0196-2892",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "10",

}

TY - JOUR

T1 - Multiple classifiers applied to multisource remote sensing data

AU - Briem, Gunnar Jakob

AU - Benediktsson, Jon Atli

AU - Sveinsson, Johannes R.

N1 - Funding Information: Manuscript received September 26, 2001; revised May 17, 2002. This research was supported in part by the Icelandic Research Council and the Research Fund of the University of Iceland. G. J. Briem was with the Department of Electrical and Computer Engineering, University of Iceland, Reykjavik IS-107, Iceland. He is now with Dimon Software, Reykjavik IS-105, Iceland. J. A. Benediktsson and J. R. Sveinsson are with the Department of Electrical and Computer Engineering, University of Iceland, Reykjavik IS-107, Iceland. Digital Object Identifier 10.1109/TGRS.2002.802476

PY - 2002/10

Y1 - 2002/10

N2 - The combination of multisource remote sensing and geographic data is believed to offer improved accuracies in land cover classification. For such classification, the conventional parametric statistical classifiers, which have been applied successfully in remote sensing for the last two decades, are not appropriate, since a convenient multivariate statistical model does not exist for the data. In this paper, several single and multiple classifiers, that are appropriate for the classification of multisource remote sensing and geographic data are considered. The focus is on multiple classifiers: bagging algorithms, boosting algorithms, and consensus-theoretic classifiers. These multiple classifiers have different characteristics. The performance of the algorithms in terms of accuracies is compared for two multisource remote sensing and geographic datasets. In the experiments, the multiple classifiers outperform the single classifiers in terms of overall accuracies.

AB - The combination of multisource remote sensing and geographic data is believed to offer improved accuracies in land cover classification. For such classification, the conventional parametric statistical classifiers, which have been applied successfully in remote sensing for the last two decades, are not appropriate, since a convenient multivariate statistical model does not exist for the data. In this paper, several single and multiple classifiers, that are appropriate for the classification of multisource remote sensing and geographic data are considered. The focus is on multiple classifiers: bagging algorithms, boosting algorithms, and consensus-theoretic classifiers. These multiple classifiers have different characteristics. The performance of the algorithms in terms of accuracies is compared for two multisource remote sensing and geographic datasets. In the experiments, the multiple classifiers outperform the single classifiers in terms of overall accuracies.

KW - Bagging

KW - Boosting

KW - Consensus theory

KW - Multiple classifiers

KW - Multisource remote sensing data

UR - http://www.scopus.com/inward/record.url?scp=0036821351&partnerID=8YFLogxK

U2 - 10.1109/TGRS.2002.802476

DO - 10.1109/TGRS.2002.802476

M3 - Article

AN - SCOPUS:0036821351

SN - 0196-2892

VL - 40

SP - 2291

EP - 2299

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

IS - 10

ER -

Multiple classifiers applied to multisource remote sensing data

Abstract

Bibliographical note

Other keywords

Access to Document

Fingerprint

Cite this