1 August 2000 Performance of an automatic target recognizer algorithm against real and two versions of synthetic imagery
Author Affiliations +
Abstract
An experiment is described in which an automatic target recognizer (ATR) algorithm is tested against real forward-looking infrared imagery and synthetic imagery. The synthetic imagery was generated by two different approaches being developed by the U.S. Army. The purpose of the experiment was to establish quantitatively the relationship between ATR performance on real and on synthetic data. The metrics for comparison are probability of detection, conditional probability of identification, and false-alarm rate.
Clarence P. Walters, Carl W. Hoover, James A. Ratches, "Performance of an automatic target recognizer algorithm against real and two versions of synthetic imagery," Optical Engineering 39(8), (1 August 2000). https://doi.org/10.1117/1.1305538 . Submission:
JOURNAL ARTICLE
6 PAGES


SHARE
Back to Top