Interferogram filtering to reduce phase noise is an essential aspect to generating useful geophysical products for both single and repeat pass interferometry. For high correlation data sets such as those obtained from single pass systems or from repeat pass data with little temporal correlation simple low pass filters such as boxcar filters often provide ample smoothing of the interferometric phase. However, in more challenging data sets where the interferometric correlation can be low, such as in repeat pass applications for deformation measurement, it is often necessary to apply more vigorous filtering to extract useful information. Goldstein-Werner filtering that weights the spectrum of an interferogram has proven effective in many applications. In this paper we extend analytic expressions for the filter performance to the case of distributed targets and multi-looking prior to filtering.
展开▼
