FLIR Operator Requirements Study
Abstract
Three experiments were carried out to determine the quantitative relationships between FLIR image quality parameters and target recognition. Simulated FLIR imagery was used in each experiment to measure recognition performance as a function of specific image quality variables. The original target imagery was produced by photographing the displayed output of an infrared sensor. A variety of target types and examples were used. In each experiment the original target images were degraded by digital image processing techniques according to specific levels of the image quality variables. Experimentation was then carried out in a laboratory situation to measure recognition performance. Data from a total of 43,764 experimental trials were collected and analyzed. The implications of the results of the three experiments to system design is discussed. The image quality variable investigated in Experiment 1 were: number of scan lines, modulation transfer function, noise, and magnification. Experiment 2 investigated the following image quality variables: number of scan lines, scan aperture size, noise, and magnification. In Experiment 3 the effects of dynamic noise on the recognition of degraded FLIR targets were investigated.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1976
- Accession Number
- ADA024288
Entities
People
- Judith M. Erickson
- Leon G. Williams
Organizations
- Honeywell International, Inc.