Author

Michael Jones

Abstract

The relationship between information capacity and the ability of photointerpreters to identify vehicles in simulated aerial reconnaissance was investigated. An aerial scene lighting simulator was constructed and used in the production of a series of simulated aerial reconnaissance photographs of models of military tanks and trucks. The information capacity of these photos was varied by defocussing the taking camera and the simulated ground scale was varied by changing the taking camera reduction. Duplicate positives of these images were evaluated by trained military photointerpreters who determined the resolving power and attempted to identify the vehicles from a key provided. The resulting empirical probability of correct identification for each vehicle was plotted against the information capacity, which was computed as one half the square of the resolving power. For a probability of correct identification of 0.80, these curves indicate that an information capacity of 7.4 bits per square meter on the ground is required to identify tanks, and 2 bits per square meter is required for identification of trucks.

Library of Congress Subject Headings

Photographic interpretation; Aerial photography

Publication Date

6-1-1978

Document Type

Thesis

Department, Program, or Center

School of Photographic Arts and Sciences (CIAS)

Advisor

Abouelata, Mohamed

Comments

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works in December 2013. Physical copy available from RIT's Wallace Library at TR810.J65

Campus

RIT – Main Campus

Share

COinS