A study of PS-InSAR method for small area urban land subsidence

Persistent Scatterers for SAR Interferometry (PS-InSAR) tecnique allows monitoring the temporal evolution of a deformation phenomenon at millimeter level, via the generation of mean deformation velocity maps and displacement time series from a data set of acquired SAR images. However, the serious noises of multi-temporal SAR images make the selection of permanent scatters (PSs) difficult. In this paper, we propose a novel PS-InSAR method based on wavelet phase analysis to solve this problem by improving the quality of interferograms. Firstly, both the real parts and the imaginary parts of SAR interferograms are processed by wavelet decomposition. Secondly, the interferograms on every level are filtered by grads dependent adaptive filter and reconstructed. Finally, interferograms are processed to identify PSs by using two indices namely, amplitude disperse index and spatial correlation index, followed by an unwrap process to extract land subsidence information. We then use this method to process 12 ENVISAT SAR imags covering Jiaxing city in Zhejiang province, China. The experimental results show its effectiveness in residuals and PSs identification and land subsidence extraction for plain urban area.