IMPROVING PHASE UNWRAPPING TECHNIQUES BY THE USE OF LOCAL FREQUENCY ESTIMATES

In multipass spaceborne synthetic aperture radar (SAR) interferometry, the two acquisitions often present low correlation levels and very no isy phase measurements that are incompatible with automatic phase unwr apping. Instead of dealing with many residues due to erroneous-wrapped phase differences, we propose to use the local frequency as measured by a spectral analysis algorithm presented in a previous paper [1], Fo r this purpose, we present two conventional unwrapping algorithms, one local and the other global, which we revisit to benefit from the robu st local frequency estimates. For a local approach based on path-follo wing techniques, we use the frequency estimates in a slope-compensated filter that extend the complex averaging up to a sufficient number of looks to eliminate residues due to the noise. Then we connect residue s due to noninterferometric features along mask components resulting f rom the detection of layovers and uncorrelated areas. For a global app roach, such as the weighted least-squares methods, we demonstrate that the use of noisy discrete phase gradient leads to a biased solution. To avoid this drawback, we propose to use the local frequency estimate and associated measure of confidence as phase gradient and weight. Re sults are presented on both topographic and differential interferogram s obtained from the ERS-1 European radar satellite over various landsc apes and the displacement field of the Landers 1992 earthquake.