Ground Deformation Mornitoring Technique Based On Time-series High Resolution SAR Images

Since differential synthetic aperture radar interferometry (D-InSAR) technologybased on repeat orbits comes out, it is applied and studied positively with the advantagesof short time interval, wide coverage and low cost. However, the conventionalD-InSAR technique is vulnerable to be affected by loss coherence of temporal andspatial and atmospheric retardation, decreases the reliability of the results. In order toovercome these constrains, some new InSAR processing methods have been proposed,the representative technologies are permanent scatterer technique and the small baselinesubset technique, what these techniques have in common is to identify and analyse thosepoint targets maintain a stable characteristic or high coherence during the time-seriesSAR images. This article aims at analyzing those stable targets on time-series and highresolution SAR images and retrieves the surface deformation in high precision.Therefore, this article does the following studies:(1) Present the basic principles and data processing processes of InSAR, DInSAR indetail, make a deep analysis of error sources which influence the accuracy of DInSAR,sum up the principles of time series DInSAR algorithms in recent years based oncoherent point target analysis.(2) Make a study of how to extract stable point target on time-series and highresolution SAR images. Make a detailed research of currently methods, such as based onamplitude, coherence and phase stability, explain the involved processing steps, pointout their advantages, disadvantages and different application situation. This articlepresent a multi-step detection method based on amplitude and phase. Finally anexperiment on high resolution TerraSAR data had been made to compare and evaluatedifferent results of each method. It verifies the advantage of multi-step detection methodbased on amplitude and phase. This method ensures the density and precision of stablepoint targets.(3) Make a study of the characteristic of stable point target on time-series highresolution SAR images, summarize the interpretation and classification of stable pointtargets.(4) Make a research of surface deformation monitoring based on time-series InSARtechnology. This paper carries out a method called Multi-Temporal InSAR, which combines the advantages of permanent scatterer technique and small baseline subsettechnique. We apply this technology and retrieve the surface subsidence along Jingjinhigh speed railway from Mar,2009to Jan,2010in Tianjin. The results show a goodagreement with historical data and verify the validity and reliability of this method.