NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
Multisensor classification of sedimentary rocksA comparison is made between linear discriminant analysis and supervised classification results based on signatures from the Landsat TM, the Thermal Infrared Multispectral Scanner (TIMS), and airborne SAR, alone and combined into extended spectral signatures for seven sedimentary rock units exposed on the margin of the Wind River Basin, Wyoming. Results from a linear discriminant analysis showed that training-area classification accuracies based on the multisensor data were improved an average of 15 percent over TM alone, 24 percent over TIMS alone, and 46 percent over SAR alone, with similar improvement resulting when supervised multisensor classification maps were compared to supervised, individual sensor classification maps. When training area signatures were used to map spectrally similar materials in an adjacent area, the average classification accuracy improved 19 percent using the multisensor data over TM alone, 2 percent over TIMS alone, and 11 percent over SAR alone. It is concluded that certain sedimentary lithologies may be accurately mapped using a single sensor, but classification of a variety of rock types can be improved using multisensor data sets that are sensitive to different characteristics such as mineralogy and surface roughness.
Document ID
19880059539
Acquisition Source
Legacy CDMS
Document Type
Reprint (Version printed in journal)
Authors
Evans, Diane
(California Institute of Technology Jet Propulsion Laboratory, Pasadena, United States)
Date Acquired
August 13, 2013
Publication Date
July 1, 1988
Publication Information
Publication: Remote Sensing of Environment
Volume: 25
ISSN: 0034-4257
Subject Category
Earth Resources And Remote Sensing
Accession Number
88A46766
Distribution Limits
Public
Copyright
Other

Available Downloads

There are no available downloads for this record.
No Preview Available