Quantitative analysis of HRR NCTR performance drivers

David Iny; Martin M. Morici

doi:10.1117/12.243073

17 June 1996 Quantitative analysis of HRR NCTR performance drivers

David Iny, Martin M. Morici

Proceedings Volume 2747, Radar Sensor Technology; (1996) https://doi.org/10.1117/12.243073
Event: Aerospace/Defense Sensing and Controls, 1996, Orlando, FL, United States

Abstract

Over the past several years Northrop Grumman has been developing Non-Cooperative Target Recognition (NCTR) technology using High Range Resolution (HRR) Radar data. Common to all NCTR efforts is the need to train classifier algorithms on limited sources of data. The classifier design must also address signature variations with aspect viewing angles and stores configurations. This paper will provide a methodology for quantifying training data segmentation issues including: (1) Degradation due to limited samples within an aspect zone; (2) Stability of scattering centers as a function of aspect angle; and (3) Stores variations. In a program supported by Wright Patterson AFB, Northrop Grumman has developed a detailed statistical model of the Airborne Radar Target Identification HRR signature data. The statistical model is based on a template alignment procedure. This model provides an analytic basis for predicting classifier performance using an associated distance metric. This paper will provide a brief discussion of our template classifier and apply the analytic model to the segmentation issues in the previous paragraph.

Citation Download Citation

David Iny and Martin M. Morici "Quantitative analysis of HRR NCTR performance drivers", Proc. SPIE 2747, Radar Sensor Technology, (17 June 1996); https://doi.org/10.1117/12.243073

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available