An evaluation of synthetic aperture radar noise reduction techniques for the smoothing of electronic speckle pattern interferometry fringes

摘要

Test objects with different surface reflection properties produce local partial decorrelation in electronic speckle pattern interferometry (ESPI) fringe patterns. In addition to this effect, the finite dimensions of the test object and illumination changes due to surface concavities or shadowing may also contribute to undesirable noise. These conditions make the analysis problems of ESPI fringe patterns similar to those encountered in synthetic aperture radar (SAR) images, where local terrain features influence the speckle statistics. This paper presents a brief review of noise reduction techniques commonly used in SAR imaging, some of which are then examined for potential application to ESPI. Computer-simulated and experimental fringe patterns are used to test SAP? noise reduction algorithms and results are compared with those obtained by means of a more conventional low-pass Fourier filter.