Perceptibility of Military Vehicle Silhouettes
Technical memo. (Final)
HUMAN ENGINEERING LAB ABERDEEN PROVING GROUND MD
Pagination or Media Count:
Experiments were conducted to assess human performance measures of the perceptibility of tracked vehicle designs as viewed through a simulated sensor system. Subjects were trained to identify silhouettes of 4 tracked and 4 wheeled military vehicles. Each targets was embedded in one of 12 levels of white noise, that is, the signal-to-noise ratio. Subjects identification responses were analyzed to produce 3 SNRs for each target at which were barely detectable, recognizable as to target type, or identification as a specific target. Analyses of the data indicated that the relative perceptibility of each target depended largely upon the number of features that it shared in common with other targets in the expected target set. Another experiment explored whether eye movement behavior could determine which target silhouette features appeared to be most critical to target discrimination. The forced choice responses of 22 subjects from the first experiment were analyzed to determine for each individual the specific SNR for detection, recognition, and identification for each of the 8 targets. Results indicated that the pattern of visual attention for the Bradley differed from the other tracked vehicles. There also did not appear to be any systematic match between strategies that subjects indicated they were using to identify targets and where they might have actually looked. These results were interpreted to mean that the pattern of visual attention is perceptually driven, whereas target identification per se is a cognitive function.
- Anatomy and Physiology
- Optical Detection and Detectors
- Combat Vehicles