A statistical model of spatial context is described and procedures for classifying remote sensing data using a context classifier are outlined. Experimental results are presented. Because the computational requirements of the context classifier are very large, its implementation on multiprocessor systems is investigated. Some of the special considerations necessary for such implementations are described, with particular reference to implementation on an array of Control Data Corporation Flexible Processors. Copyright © 1980 by The Institute of Electrical ind Electronics Engineers, Inc.
