IEEE 2002 Radar Conference April 22-25 in Long Beach, CA

~~Ordering Copies of Tutorial Notes (PDF 12 kB)~~

Ordering Bistatic Radar Book by Willis (PDF 24 kB)

4.2 Automatic Target Recognition Using SAR

Instructor: Dr. Les Novak - Lincoln Labs

Thursday April 25, 1:30 p.m. to 5:30 p.m.

Abstract

This tutorial on Automatic Target Recognition (ATR) using Synthetic Aperture Radar (SAR) imagery focuses on two main areas of research: (1) SAR ATR using fully polarimetric data and (2) SAR ATR using superresolution image processing. In part 1 of the tutorial we describe a fully polarimetric Ka-band SAR sensor, and we develop various techniques, including the polarimetric whitening filter (PWF) and the polarimetric matched filter (PMF), for optimum processing of the fully polarimetric data. A complete end-to-end target recognition system is developed, and performance results versus resolution and polarization are summarized. In part 2 of the tutorial, various superresolution processing algorithms applied to SAR data are described, and for each algorithm the improvement in ATR performance is quantified. Recognition performance results are presented for a 20-target classifier, and lessons learned from extensive SAR ATR evaluations conducted at MIT Lincoln Laboratory are summarized.

Additional ATR topics to be presented include: target recognition performance comparisons using 1-D high range resolution (HRR) profile classifiers versus 2-D SAR classifiers; model-based classifier performance; combined polarimetric/superresolution processing; and low frequency foliage penetration (FOPEN) SAR target detection and discrimination.

Bio

Dr. Leslie Novak received his PhD from UCLA. He is a Fellow of the IEEE and has been at MIT Lincoln Laboratory since 1977. He currently holds the position of Senior Staff in Division 4 and is assigned to the Sensor Exploitation Group, where he is involved in the development of target detection, acquisition, and classification algorithms for SAR (synthetic aperture radar) systems. Recently, he has been performing studies on polarimetric radar signal processing algorithms and superresolution signal processing algorithms for synthetic aperture radar systems. He has published numerous papers on optimum processing of polarimetric radar data (including the polarimetric whitening filter and the polarimetric matched filter) and on SAR superresolution processing applied to the radar automatic target recognition problem.