Fuzzy logic approach for computing the probability of target detection in cluttered scenes

A tremendous number of fuzzy system-based applications are currently available. A new application of fuzzy systems theory is presented to compute the probability of target detection in cluttered infrared images. Both the Mamdani and Assilian and Sugeno's fuzzy-based models have been investigate...

Full description

Saved in:
Bibliographic Details
Published in:Optical Engineering 1996-12, Vol.35 (12), p.3623-3636
Main Authors: Meitzler, Thomas J, Arafeh, Labib, Singh, Harpreet, Gerhart, Grant R
Format: Article
Language:English
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A tremendous number of fuzzy system-based applications are currently available. A new application of fuzzy systems theory is presented to compute the probability of target detection in cluttered infrared images. Both the Mamdani and Assilian and Sugeno's fuzzy-based models have been investigated. A data set for one vehicle has been used to model the relationships between three input parameters, the Der clutter metric, the probability of edge (POE) metric, the Schmieder variance-based clutter metric, and the experimental probability of detection value. Both fuzzy models have given excellent predictive probability of detection values that are highly correlated to the observed probability of detection values. The preliminary results obtained suggest the adequacy of fuzzy-based modeling techniques to model the perception of targets in infrared and visual scenes. ©
ISSN:0091-3286
1560-2303
DOI:10.1117/1.601081