Conjugate-gradient neural networks in classification of multisource and very-high-dimensional remote sensing data

J. A. Benediktsson; P. H. Swain; O. K. Ersoy

doi:10.1080/01431169308904316

Conjugate-gradient neural networks in classification of multisource and very-high-dimensional remote sensing data

J. A. Benediktsson, P. H. Swain, O. K. Ersoy

University of Iceland

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

89 Citations (Scopus)

Abstract

Application of neural networks to classification of remote sensing data is discussed. Conventional two-layer backpropagation is found to give good results in classification of remote sensing data but is not efficient in training. A more efficient variant, based on conjugate-gradient optimization, is used for classification of multisourcc remote sensing and geographic data and very-highdimensional data. The conjugate-gradient neural networks give excellent performance in classification of multisource data but do not compare as well with statistical methods in classification of very-high-dimcnsional data.

Original language	English
Pages (from-to)	2883-2903
Number of pages	21
Journal	International Journal of Remote Sensing
Volume	14
Issue number	15
DOIs	https://doi.org/10.1080/01431169308904316
Publication status	Published - Oct 1993

Bibliographical note

Funding Information:
This research was supported in part by the National Aeronautics and Space Administration (NASA) through Grant No. NAGW-925, the Icelandic Council of Science and the Research Fund of the University of Iceland.

Access to Document

10.1080/01431169308904316

Cite this

@article{39b6ddb97bc54eaaa2581cc1abb56a4d,

title = "Conjugate-gradient neural networks in classification of multisource and very-high-dimensional remote sensing data",

abstract = "Application of neural networks to classification of remote sensing data is discussed. Conventional two-layer backpropagation is found to give good results in classification of remote sensing data but is not efficient in training. A more efficient variant, based on conjugate-gradient optimization, is used for classification of multisourcc remote sensing and geographic data and very-highdimensional data. The conjugate-gradient neural networks give excellent performance in classification of multisource data but do not compare as well with statistical methods in classification of very-high-dimcnsional data.",

author = "Benediktsson, {J. A.} and Swain, {P. H.} and Ersoy, {O. K.}",

note = "Funding Information: This research was supported in part by the National Aeronautics and Space Administration (NASA) through Grant No. NAGW-925, the Icelandic Council of Science and the Research Fund of the University of Iceland.",

year = "1993",

month = oct,

doi = "10.1080/01431169308904316",

language = "English",

volume = "14",

pages = "2883--2903",

journal = "International Journal of Remote Sensing",

issn = "0143-1161",

publisher = "Taylor and Francis Ltd.",

number = "15",

}

TY - JOUR

T1 - Conjugate-gradient neural networks in classification of multisource and very-high-dimensional remote sensing data

AU - Benediktsson, J. A.

AU - Swain, P. H.

AU - Ersoy, O. K.

N1 - Funding Information: This research was supported in part by the National Aeronautics and Space Administration (NASA) through Grant No. NAGW-925, the Icelandic Council of Science and the Research Fund of the University of Iceland.

PY - 1993/10

Y1 - 1993/10

N2 - Application of neural networks to classification of remote sensing data is discussed. Conventional two-layer backpropagation is found to give good results in classification of remote sensing data but is not efficient in training. A more efficient variant, based on conjugate-gradient optimization, is used for classification of multisourcc remote sensing and geographic data and very-highdimensional data. The conjugate-gradient neural networks give excellent performance in classification of multisource data but do not compare as well with statistical methods in classification of very-high-dimcnsional data.

AB - Application of neural networks to classification of remote sensing data is discussed. Conventional two-layer backpropagation is found to give good results in classification of remote sensing data but is not efficient in training. A more efficient variant, based on conjugate-gradient optimization, is used for classification of multisourcc remote sensing and geographic data and very-highdimensional data. The conjugate-gradient neural networks give excellent performance in classification of multisource data but do not compare as well with statistical methods in classification of very-high-dimcnsional data.

UR - http://www.scopus.com/inward/record.url?scp=0027789411&partnerID=8YFLogxK

U2 - 10.1080/01431169308904316

DO - 10.1080/01431169308904316

M3 - Article

AN - SCOPUS:0027789411

SN - 0143-1161

VL - 14

SP - 2883

EP - 2903

JO - International Journal of Remote Sensing

JF - International Journal of Remote Sensing

IS - 15

ER -

Conjugate-gradient neural networks in classification of multisource and very-high-dimensional remote sensing data

Abstract

Bibliographical note

Access to Document

Fingerprint

Cite this