Neural Network Approaches Versus Statistical Methods in Classification of Multisource Remote Sensing Data

Jon A. Benediktsson; Philip H. Swain; Okan K. Ersoy

doi:10.1109/TGRS.1990.572944

Neural Network Approaches Versus Statistical Methods in Classification of Multisource Remote Sensing Data

Jon A. Benediktsson
, Philip H. Swain
, Okan K. Ersoy

University of Iceland

University of Iceland

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

Abstract

Neural network learning procedures and statistical classification methods are applied and compared empirically in classification of multisource remote sensing and geographic data. Statistical multisource classification by means of a method based on Bayesian classification theory is also investigated and modified. The modifications permit control of the influence of the data sources involved in the classification process. Reliability measures are introduced to rank the quality of the data sources. The data sources are then weighted according to these rankings in the statistical multisource classification. Four data sources are used in experiments: Landsat MSS data and three forms of topographic data (elevation, slope, and aspect). Experimental results show that the two different approaches have unique advantages and disadvantages in this classification application.

Original language	English
Pages (from-to)	540-552
Number of pages	13
Journal	IEEE Transactions on Geoscience and Remote Sensing
Volume	28
Issue number	4
DOIs	https://doi.org/10.1109/TGRS.1990.572944
Publication status	Published - Jul 1990

Bibliographical note

Funding Information: Manuscript received October 8, 1989, revised March 1, 1990 This work was supported in part by the National Aeronautics and Space Administration under contract NAGW 925 The authors are with the School of Electrical Engineering and the Laboratory for Applications of Remote Sensing, Purdue University, West La fayette, IN 47907 IEEE Log Number 9036095

Access to Document

10.1109/TGRS.1990.572944

Cite this

@article{69e982d745004343ab6fe449482e30ec,

title = "Neural Network Approaches Versus Statistical Methods in Classification of Multisource Remote Sensing Data",

abstract = "Neural network learning procedures and statistical classification methods are applied and compared empirically in classification of multisource remote sensing and geographic data. Statistical multisource classification by means of a method based on Bayesian classification theory is also investigated and modified. The modifications permit control of the influence of the data sources involved in the classification process. Reliability measures are introduced to rank the quality of the data sources. The data sources are then weighted according to these rankings in the statistical multisource classification. Four data sources are used in experiments: Landsat MSS data and three forms of topographic data (elevation, slope, and aspect). Experimental results show that the two different approaches have unique advantages and disadvantages in this classification application.",

author = "Benediktsson, \{Jon A.\} and Swain, \{Philip H.\} and Ersoy, \{Okan K.\}",

note = "Funding Information: Manuscript received October 8, 1989, revised March 1, 1990 This work was supported in part by the National Aeronautics and Space Administration under contract NAGW 925 The authors are with the School of Electrical Engineering and the Laboratory for Applications of Remote Sensing, Purdue University, West La fayette, IN 47907 IEEE Log Number 9036095",

year = "1990",

month = jul,

doi = "10.1109/TGRS.1990.572944",

language = "English",

volume = "28",

pages = "540--552",

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

issn = "0196-2892",

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

number = "4",

}

TY - JOUR

T1 - Neural Network Approaches Versus Statistical Methods in Classification of Multisource Remote Sensing Data

AU - Benediktsson, Jon A.

AU - Swain, Philip H.

AU - Ersoy, Okan K.

N1 - Funding Information: Manuscript received October 8, 1989, revised March 1, 1990 This work was supported in part by the National Aeronautics and Space Administration under contract NAGW 925 The authors are with the School of Electrical Engineering and the Laboratory for Applications of Remote Sensing, Purdue University, West La fayette, IN 47907 IEEE Log Number 9036095

PY - 1990/7

Y1 - 1990/7

N2 - Neural network learning procedures and statistical classification methods are applied and compared empirically in classification of multisource remote sensing and geographic data. Statistical multisource classification by means of a method based on Bayesian classification theory is also investigated and modified. The modifications permit control of the influence of the data sources involved in the classification process. Reliability measures are introduced to rank the quality of the data sources. The data sources are then weighted according to these rankings in the statistical multisource classification. Four data sources are used in experiments: Landsat MSS data and three forms of topographic data (elevation, slope, and aspect). Experimental results show that the two different approaches have unique advantages and disadvantages in this classification application.

AB - Neural network learning procedures and statistical classification methods are applied and compared empirically in classification of multisource remote sensing and geographic data. Statistical multisource classification by means of a method based on Bayesian classification theory is also investigated and modified. The modifications permit control of the influence of the data sources involved in the classification process. Reliability measures are introduced to rank the quality of the data sources. The data sources are then weighted according to these rankings in the statistical multisource classification. Four data sources are used in experiments: Landsat MSS data and three forms of topographic data (elevation, slope, and aspect). Experimental results show that the two different approaches have unique advantages and disadvantages in this classification application.

UR - https://www.scopus.com/pages/publications/0025453627

U2 - 10.1109/TGRS.1990.572944

DO - 10.1109/TGRS.1990.572944

M3 - Article

SN - 0196-2892

VL - 28

SP - 540

EP - 552

JO - IEEE Transactions on Geoscience and Remote Sensing

JF - IEEE Transactions on Geoscience and Remote Sensing

IS - 4

ER -

Neural Network Approaches Versus Statistical Methods in Classification of Multisource Remote Sensing Data

Abstract

Bibliographical note

Access to Document

Fingerprint

Cite this