Adaptive Synthetic Aperture Radar (SAR) Imaging for Optimal Cross-Range Resolution and Image Quality in NDE Applications

Abstract

An adaptive scanning algorithm is proposed for reducing scanning time associated with generating synthetic aperture radar (SAR) images, for nondestructive evaluation (NDE) applications, while preserving optimal cross-range resolution and image quality. The proposed method is based on the fact that optimum SAR cross-range resolution is target depth, synthetic aperture dimension, and scanning step size dependent. Hence, in the proposed method, the target properties, such as position and depth, are first estimated using an initial coarse scan of the scene of interest. Then, a localized dense scan is performed for obtaining optimal spatial resolution and image quality. During this procedure, the synthetic aperture dimension of the localized dense scan is adaptively increased (optimized) by reducing pixel-by-pixel image root-mean-square (rms) intensity difference between the images produced before and after each increment increase in the synthetic aperture dimension of the localized dense scan. At the same time, scanning step size is determined in a manner, which is related to the target depth in order to reduce scanning time and to insure obtaining (close to) optimal spatial sampling. Results of several illustrative measurements are provided by which the efficacy of the proposed method is demonstrated. The proposed algorithm, which mainly applies for imaging a localized flaw or a strong-scattering target in a relatively homogenous environment, is common in NDE applications and can be implemented using an automated and adaptive process.