Improving Goldstein filter by image entropy for InSAR interferogram filtering

InSAR has proven to be a viable method for measuring surface topography. Interferometric phase filtering is an important method to improve the quality of DEM in InSAR data processing. We present a modification to the adaptive Goldstein radar interferogram filter which improves the quality of interferometry. The proposed approach combines correlation coefficient and image entropy instead of the filter parameter α in Goldstein filter algorithm. This modification enhances ability to identify the level of image noise and adopts more appropriate Goldstein filter parameter.