Automated adaptation for ATR algorithms

Peter F. Symosek, Michael E. Bazakos

Research output: Contribution to journalConference articlepeer-review

Abstract

For an Automatic Target Recognition (ATR) technology contract, sponsored by the United States Marine Corps Systems Command and by Coastal Systems Station (COASTSYSTA), Honeywell designed, mapped to Khoros, and evaluated state-ofthe-art algorithms for target discrimination from an airborne platform. Honeywell's baseline approach to improve traditional algorithm robustness is to use a functional maximization approach for representations of algorithm performance as a function of image metrics and algorithm parameters. Revised ATR parameter values are established by a hillclimbing algorithm that revises the ATR algorithm parameter values in the direction of the largest gradient of the function, thus attaining improved performance for a greater variety of scenarios than those for which the system was trained. The baseline ATR algorithms implemented for this program are designed to effectively exploit spectral features to enhance target cueing reliability. An innovative approach for the mapping of three of the individual waveband images from an array of multispectral images into a feature map which obtains high target vs. background contrast is discussed. Experimental results are shown for flight test imagery.

Original languageEnglish (US)
Pages (from-to)724-735
Number of pages12
JournalProceedings of SPIE - The International Society for Optical Engineering
Volume2496
DOIs
StatePublished - Jun 20 1995
Externally publishedYes
EventDetection Technologies for Mines and Minelike Targets 1995 - Orlando, United States
Duration: Apr 17 1995Apr 21 1995

Keywords

  • Automatic Target Recognition
  • Mine detection
  • Multispectral image processing
  • Parameter adaptation

Fingerprint Dive into the research topics of 'Automated adaptation for ATR algorithms'. Together they form a unique fingerprint.

Cite this