NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
Conjugate-Gradient Neural Networks in Classification of Multisource and Very-High-Dimensional Remote Sensing DataApplication 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 multisource remote sensing and geographic data and very-high-dimensional 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-dimentional data.
Document ID
19970022430
Document Type
Reprint (Version printed in journal)
Authors
Benediktsson, J. A. (Iceland Univ. Reykjavik, Iceland)
Swain, P. H. (Purdue Univ. West Lafayette, IN United States)
Ersoy, O. K. (Purdue Univ. West Lafayette, IN United States)
Date Acquired
August 17, 2013
Publication Date
January 21, 1993
Publication Information
Publication: International Journal of Remote Sensing
Volume: 14
Issue: 15
ISSN: 0143-1161
Subject Category
Earth Resources and Remote Sensing
Report/Patent Number
NAS 1.26:204335
NASA-CR-204335
Funding Number(s)
CONTRACT_GRANT: NAGw-925
Distribution Limits
Public
Copyright
Public Use Permitted.
Document Inquiry