1 September 1997 Complete automatic target cuer/recognition system for tactical forward-looking infrared images
Author Affiliations +
Abstract
A complete forward-looking IR (FLIR) automatic target cuer/ recognizer (ATC/R) is presented. The data used for development and testing of this ATC/R are first generation FLIR images collected using a F-15E. The database contains thousands of images with various mission profiles and target arrangements. The specific target of interest is a mobile missile launcher, the primary target. The goal is to locate all vehicles (secondary targets) within a scene and identify the primary targets. The system developed and tested includes an image segmenter, region cluster algorithm, feature extractor, and classifier. Conventional image processing algorithms in conjunction with neural network techniques are used to form a complete ATC/R system. The conventional techniques include hit/miss filtering, difference of Gaussian filtering, and region clustering. A neural network (multilayer perceptron) is used for classification. These algorithms are developed, tested and then combined into a functional ATC/R system. Overall primary target detection rate (cuer) is 84% with a 69% primary target identification (recognizer) rate at ranges relevant to munitions release. Furthermore, the false alarm rate (a nontarget cued as a target) is only 2.3 per scene. The research is being completed with a 10 flight test profile using third generation FLIR images.
Brian E. Ernisse, Brian E. Ernisse, Steven K. Rogers, Steven K. Rogers, Martin P. DeSimio, Martin P. DeSimio, Richard A. Raines, Richard A. Raines, } "Complete automatic target cuer/recognition system for tactical forward-looking infrared images," Optical Engineering 36(9), (1 September 1997). https://doi.org/10.1117/1.601484 . Submission:
JOURNAL ARTICLE
11 PAGES


SHARE
Back to Top