Multi-Polarization Methods to Detect Damages Related to Earthquakes

In this study, multi-polarimetric features extracted from dual-polarimetric (DP) Synthetic Aperture Radar (SAR) data collected by the Sentinel-l SAR mission are exploited to analyze damages due to earthquakes. First, conventional single-polarization features, based on the coherence between pre- and post-event imagery, are analyzed using both the co-polarized and the cross-polarized channels to point out that the information carried on the two channels is different. Then, a coherent dual-polarimetric feature, based on the inter-channel coherence, is proposed. Results show that the latter feature allows to detect the changes after an earthquake using only two SAR acquisitions, i.e.; before and after the earthquake. Finally, a change detection algorithm based on the eigenvalues of the difference of covariance matrices, is proposed. Experimental results show that the DP information allows to improve the results obtained by conventional single polarization change detection techniques.