Share Email Print

Proceedings Paper

Method for reducing dimensionality in ATR systems
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

A method for robustly selecting reduced dimension statistics for pattern recognition systems is described. A stochastic model for each target or object is assumed parameterized by a finite dimensional vector. Data and parameter vectors are assumed to be long. As the size of these vectors increases, the performance improves to a point and then degrades; this trend is called the peaking phenomenon. A new, more robust method for selecting reduced dimension approximations is presented. This method selects variables if a measure of the amount of information provided exceeds a given level. This method is applied to distributions in the exponential family, performance is compared to other methods, and an analytical expression for performance is asymptotically approximated. In all cases studied, performance is better than with other known methods.

Paper Details

Date Published: 17 August 2000
PDF: 12 pages
Proc. SPIE 4050, Automatic Target Recognition X, (17 August 2000); doi: 10.1117/12.395582
Show Author Affiliations
Joseph A. O'Sullivan, Washington Univ. (United States)
Natalia A. Schmid, Washington Univ. (United States)

Published in SPIE Proceedings Vol. 4050:
Automatic Target Recognition X
Firooz A. Sadjadi, Editor(s)

© SPIE. Terms of Use
Back to Top