Vehicle Detection on Television: A Laboratory Experiment.
Abstract
An experiment was conducted to measure the ability of observers to search for and detect vehicular targets in a real-world scene displayed on a TV monitor. Search time was limited to 3 seconds; the variables investigated were (1) angular subtense of the display and target, (2) the number of TV scan lines making up the target, and (3) the target-background contrast. The angular subtense of the display and target significantly affected performance; it was found that the display should subtend at least 8 and preferably 10 degrees to the obsever. Larger displays yielded little improvement in performance. The number of TV scan lines making up the target did not significantly affect performance; the lowest number used in the tests(seven) was sufficient, and increasing the number did not improve performance. Target location in the scene significantly affected performance; targets located in the bottom of the scene were easier to detect. Because of the oblique view, the targets in the bottom were larger and there was less terrain/foliage clutter. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1974
- Accession Number
- AD0919898
Entities
People
- George L. Craig
Organizations
- Naval Air Weapons Station China Lake