Change Detection in Aerial Photo Coverage as Influenced by Methods of Comparison

Carl L. Klingberg · Charles Elworth · Sidney Epstein

Jan 1969 · U.S. Army Behavioral Science Research Laboratory

About this ebook

A study was made to investigate the effects of selected variations in procedure--methods of comparing current with prior imagery cover, familiarity of the interpreter with the area, and time allowed for interpretation--on the completeness and accuracy with which change in the status of targets is detected. Under the 5-minute time limit, the computer comparison of independent interpretations of current and prior photo coverage resulted in more complete change detection than did the other methods employed and about the same degree of accuracy. The advantage of computer comparison was less marked under the 15- minute time limit. Providing the interpreter with overlay plus a written report on the prior imagery was the least effective of the comparison methods investigated. Without the actual imagery from the earlier mission, the interpreter has no way of correcting errors in the earlier interpretation. The method also was time-consuming. Interpreter familiarity with the terrain through preparation of reports on the prior imagery resulted in more complete and more accurate change detection under the 5-minute time limit. Under the 15- minute time limit, the advantage was negligible.