Phase unwrapping of SAR interferograms after wavelet denoising

Obtaining a digital elevation model (DEM) with the help of synthetic aperture radar (SAR) interferometry requires 2D phase unwrapping known to be sensitive to noise. Especially the existence of local inconsistencies of the data, referred to as residues, slows down the convergence of typical unwrapping algorithms and introduces inaccuracies into the computed results, affecting directly terrain height inversion. This suggests applying the computationally efficient Donoho wavelet denoising technique with the complex interferogram as input, preserving the lines of phase discontinuity (fringes) and yielding filtered images of good visual quality. The authors examine the quantitative effects of this pre-processing.