@article{oai:meio-u.repo.nii.ac.jp:00001131,
 author = {ARATHOON, Peter and KYAN, Seiki and TANIGUCHI, Yuuji},
 issue = {33},
 month = {2020-12-03},
 note = {Landsat TM data for Okinawa is analysed by means of training areas representing 8 different types of ground cover which are cloud, water, earth, houses, farm (vegetation other than trees), tarmac (includes concrete), trees and shadow. Statistics for each area are evaluated and output to NLP in the form of average tables, matrices, spectra, level slices and two-dimensional band slices. The original data may also be output to the NLP, (rather than to specialized colour output hardware, as is the case in much more elaborate systems). One band's data is reproduced in eight shades of grey representing relative intensity. Two classification procedures (6-dimensional minimum distance and 6-dimensional maximum likelihood) are applied, and their results compared. Image output for these is similar to that for original data, except with patterns rather than halftones representing different classes., application/pdf},
 pages = {45--60},
 title = {OUTPUT OF LANDSAT ANALYSIS TO NLP AND CLASSIFICATION OF DATA FOR OKINAWA},
 year = {}
}