NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Back to Results
A Spatial Classifier for Multispectral Data Using Contextual InformationConnectivity describes the spatial relationship among pixels. A spatial classifier which employs the sigma probability concept of the Gaussian distribution and a type of contextual information connectivity of the pixels, is studied in this paper. This spatial classifier attempts to replicate the kind of spatial synthesis done by the human analyst during visual interpretation or to capture the spatial relationships inherent in an aerial photograph. Several classification results of the Landsat TM data using this classifier with different window sizes for capturing the contextual information are illustrated and compared.
Document ID
20000032231
Acquisition Source
Headquarters
Document Type
Conference Paper
Authors
Hung, Chih-Cheng
(Alabama A & M Univ. Normal, AL United States)
Fahsi, Ahmed
(Alabama A & M Univ. Normal, AL United States)
Coleman, Tommy
(Alabama A & M Univ. Normal, AL United States)
Date Acquired
August 19, 2013
Publication Date
February 22, 1998
Publication Information
Publication: NASA University Research Centers Technical Advances in Aeronautics, Space Sciences and Technology, Earth Systems Sciences, Global Hydrology, and Education
Volume: 2 and 3
Subject Category
Documentation And Information Science
Funding Number(s)
CONTRACT_GRANT: SUB90-206-A12
CONTRACT_GRANT: NCCW-84
Distribution Limits
Public
Copyright
Work of the US Gov. Public Use Permitted.
Document Inquiry

Available Downloads

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