A novel application of Kalman Filter: 3D Reconstruction of urban areas from InSAR data

The three dimensional (3D) reconstruction of urban areas using Interferometric Synthetic Aperture Radar (InSAR) systems can be particularly difficult to be carried out due to presence of high interferometric phase jumps. These discontinuities, generated from the presence of man-made structures, make the unwrapping problem, needed for the 3D reconstruction non trivial. In this paper we propose an approach for phase unwrapping, and consequently for 3D reconstruction, that exploits multiple acquired interferograms and combines them using the Extended Kalman Filter (EKF). Differently form other EKF based approach phase unwrapping algorithms, the proposed technique is able to take into account and correctly solving height discontinuities, making the algorithm particularly suitable for 3D reconstruction in urban areas. The methodology has been tested on real high resolution datasets, showing interesting results.