Target-Centered Models and Information-Theoretic Segmentation for Automatic Target Recognition

Authors:
Michael D. Devore;Joseph A. O'sullivan
Affiliations:
Electronic Systems and Signals Research Laboratory, Dept. of Electrical Engineering, Washington University, St. Louis, MO 63130 mdd2@cis.wustl.edu;Electronic Systems and Signals Research Laboratory, Dept. of Electrical Engineering, Washington University, St. Louis, MO 63130, jao@ee.wustl.edu
Venue:
Multidimensional Systems and Signal Processing
Year:
2003

Citing 5
Cited 2

Hilbert-Schmidt Lower Bounds for Estimators on Matrix Lie Groups for ATR

IEEE Transactions on Pattern Analysis and Machine Intelligence
Automatic target recognition using high-resolution radar range-profiles

Automatic target recognition using high-resolution radar range-profiles
Information-theoretic image formation

IEEE Transactions on Information Theory
Model-based classification of radar images

IEEE Transactions on Information Theory
Thresholding method for dimensionality reduction in recognition systems

IEEE Transactions on Information Theory

Estimates of Error Probability for Complex Gaussian Channels with Generalized Likelihood Ratio Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Automatic target recognition using waveform diversity in radar sensor networks

Pattern Recognition Letters

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present an approach to automatic target recognition (ATR) from synthetic aperture radar (SAR) imagery which combines advantages of both model-based and template-based approaches. Prior observations are used to estimate the statistical properties of reflectance over regions in the training scene. These target-centered statistical models can then be used to estimate the statistical properties of sensor output for arbitrary pose. Two-sided hypothesis tests which are maximally powerful at the most likely alternative are developed in a information-theoretic framework to address target model segmentation and confuser rejection. Segmentation of target from clutter is performed in the target-centered coordinate system using all prior observations to produce a consistent segmentation over all poses. We present performance and computation complexity results as a function of segmentation threshold, confuser-rejection threshold, and operating conditions for publicly available SAR data.