Share Email Print
cover

Proceedings Paper

Robust SAR ATR via set-valued classifiers: new results
Author(s): John R. Hoffman; Ronald P. S. Mahler; Ravi B. Ravichandran; Raman K. Mehra; Stanton H. Musick
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

“Robust identification” in SAR ATR refers to the problem of determining target identity despite the confounding effects of “extended operating conditions” (EOCs). EOC’s are statistically uncharacterizable SAR intensity-signature variations caused by mud, dents, turret articulations, etc. This paper describes a robust ATR approach based on the idea of (1) hedging against EOCs by attaching “random error bars” (random intervals) to each value of the image likelihood function; (2) constructing a “generalized likelihood function” from them; and (3) using a set-valued, MLE-like approach to robustly estimate target type. We compare three such classifiers, showing that they outperform conventional approaches under EOC conditions.

Paper Details

Date Published: 25 August 2003
PDF: 12 pages
Proc. SPIE 5096, Signal Processing, Sensor Fusion, and Target Recognition XII, (25 August 2003); doi: 10.1117/12.488540
Show Author Affiliations
John R. Hoffman, Lockheed Martin Tactical Systems (United States)
Ronald P. S. Mahler, Lockheed Martin Tactical Systems (United States)
Ravi B. Ravichandran, Scientific Systems Co., Inc. (United States)
Raman K. Mehra, Scientific Systems Co., Inc. (United States)
Stanton H. Musick, Air Force Research Lab. (United States)


Published in SPIE Proceedings Vol. 5096:
Signal Processing, Sensor Fusion, and Target Recognition XII
Ivan Kadar, Editor(s)

© SPIE. Terms of Use
Back to Top