Controlling Contrast in Target Acquisition Simulations Involving Complex Backgrounds

Abstract

Visual target acquisition (TA) often involves detecting targets against natural backgrounds that have complex luminance distributions. The purpose of this study was to evaluate a simple technique that controls target contrast in the presence of varying backgrounds. Target contrast was measured by the root mean square (rms) method and was controlled by adjusting only the target luminance, leaving the background unchanged. The technique was tested in a TA paradigm in which observers searched for an aircraft that was embedded in 1) a uniform background, 2) a natural terrain background, or 3) a moving natural terrain background. Four target contrast levels were tested. The results showed that TA time varied with background and target contrast. Significant differences in TA time were observed among the different backgrounds for targets of the same physical contrast, especially at low contrast levels. Although contrast had a systematic effect on TA performance, factors other than contrast influenced TA performance. It was concluded that background structure increased TA time by camouflaging targets and by introducing distractors to the task. Such an approach could be used to model TA performance under conditions where target and background complexity are an inherent feature of the TA task.