Share Email Print
cover

Proceedings Paper

Predicting the performance of matched filter image detectors for non-stationary scenes and noise
Author(s): Robert E. Stovall
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The performance of matched filter scene detectors has been widely studied and methods for predicting its performance are available under assumptions of scene and noise stationarity. In this paper correlator performance is investigated for the more common situation when stationarity cannot be assumed. Sensor noise models are derived and combined with two binary and gray scale correlator implementations to yield an expression for the expected performance of a given system for a specific scene. This expression is implemented in an algorithm that is demonstrated on real-world imagery using fairly simple image processor hardware.

Paper Details

Date Published: 9 January 1984
PDF: 8 pages
Proc. SPIE 0432, Applications of Digital Image Processing VI, (9 January 1984); doi: 10.1117/12.936654
Show Author Affiliations
Robert E. Stovall, Defense Advanced Research Projects Agency (United States)


Published in SPIE Proceedings Vol. 0432:
Applications of Digital Image Processing VI
Andrew G. Tescher, Editor(s)

© SPIE. Terms of Use
Back to Top