Abstract
An experiment was conducted to measure observers' performance in detecting military targets in structured scenes with a high density of man-made features, i.e., “urban clutter”. The scenes were simulations of those produced by an infrared (IR) imaging system in air-to-ground situations. Scenes were generated with various signal-to-clutter ratios (SCR's), and were filtered to produce various levels of resolution. Detection performance was measured using a rating-scale detection task. Sensitivity (d1) increased with resolution, but varied little with SCR. Contrary to expectation, detection performance for a given level of resolution and SCR was better in the present urban scenes than in rural scenes used in a previous study. The findings help define requirements for the design and employment of IR imaging systems, and also suggest directions for future research directed at better understanding target detection processes in structured backgrounds.
Get full access to this article
View all access options for this article.