Multibaseline SAR Interferometry for Automatic DEM Reconstruction

Multibaseline SAR interferometry can be successfully exploited for automatic phase unwrapping and high quality Digital Elevation Model (DEM) reconstruction. The information coming from several interferograms with different baselines increases the elevation ambiguity interval and allows automatic phase unwrapping. The height of each pixel in the image is considered as a random variable: a Maximum Likelihood (ML) estimation of the height is carried out by exploiting the probability density function of the interferometric phase, that depends on the local coherence value. Atmospheric effects are taken into account and considered as a low frequency distortion (with correlation length greater than 1-2 km) of the phase to height conversion function for each interferogram. After the phase unwrapping step is possible to combine all the information available getting a combined DEM that is more reliable than each single DEM. Finally, using multiple images is possible to produce high resolution coherence maps (i.e. an ensemble average is used instead of a space average) that show what actually remains unchanged within all images. Results obtained using ERS-1 data of the area of Bonn (3 days repeat cycle) and Tandem data of two italian volcanos (Vesuvius and Etna) are presented