Improved Digital Elevation Model creation using SAR Interferometry in plane and undulating terrains

The Digital Elevation Model (DEM), which represents the variation of elevation in a terrain at spatial level, is an important source of input to a variety of applications for deriving a number of terrain parameters such as relative relief, slope, aspect direction, profiles and cross-sections, earthwork and volume computations etc. In recent years, Synthetic Aperture Radar Interferometry (InSAR) has been viewed as a powerful approach to derive quality DEMs from a pair of SAR images. In this paper, the usefulness of the technique has been demonstrated for the generation of DEMs of two markedly different terrains, viz, plane and undulating. As the interferometry is often limited by the decorelation effects from thermal noise, temporal changes and baseline geometry, an adaptive filtering algorithm has been implemented and its effect assessed in reducing the phase noise and in improving the accuracy of phase unwrapping. The quality of the DEMs has been evaluated with known mean sea level elevations as derived from the topographical maps of the study areas. In plane areas, the RMS errors between the InSAR derived elevations and the map derived elevations was obtained as 7.9 m, which further improved to 6.2 m on application of adaptive filtering. Whereas, the RMSE values in case of undulating terrain was found to be 7.2 m on application of adaptive filtering. These results clearly demonstrate the usefulness of InSAR technique for generation of DEM at meter level accuracy, which is sufficient for many engineering applications.