Differential SAR Interferometry For Deformation Monitoring Of Geological Disaster

China is a country prone to geological disasters, geological disasters each year to the state not only caused huge economic and social impact, also caused a great life and property losses. Geological disasters are often accompanied by surface deformation, so the surface deformation monitoring is the prevention of geological disasters, reduce disaster losses, rapid recovery of important post-disaster reconstruction. There are various means of monitoring the surface deformation, including Synthetic Aperture Radar differential interferometry (D-InSAR) technology. Recently, for its wide monitoring area, work around the clock, quick operation, low cost, continuous deformation monitoring advantages, D-InSAR becomes one of the most important tools in monitoring of surface deformation. Related research and application has achieved fruitful results.At present, D-InSAR technique, which developed from conventional deformation monitoring by use of only two images achieved from different time to update monitoring long time slow deformation with sequence of SAR data, further improve the detection of geological disaster monitoring capabilities. But the SAR Differential Interferometry with sequence of data need to resolve the time and space decorrelation problem and eliminate the delay impact of the atmosphere at different times, to overcome these shortcomings, recently developed sequence differential SAR interferometry based on coherent point target get more and more attention, and now becomes an important aspect of D-InSAR technique and application. Through analysis and summary the development and the actual needs of current objective status, sequence differential SAR interferometry based on the coherent point to monitor deformation in the geological disasters are also facing some problem and needs further study to resolve:first, InSAR image registration in rough area, especially deregistration caused by the phase and amplitude mismatch requires further research; Second, sequential SAR interference optimal selection problems; Furthermore, Insufficient research on sequence differential SAR interferometry based on coherent point target apply to different types of geological deformation monitoring. To solve these problems, this paper studied the following aspects:1) InSAR image registration based on local topography feature. Firstly InSAR image conventional registration methods are analyzed and summarized, then without terrain information aids and terrain information based InSAR multi-level matching method are studied and compared, and two pieces of InSAR data featured with rough topography in Three Gorges area are used to experiment and validate of this two kinds of algorithms. In addition, from the registration point offset, the correlation coefficient, the distribution of correlation coefficient aspects, these two algorithms were compared. The paper analyzes the effects of terrain factors on the SAR imaging and InSAR coherence, and focuses on the amplitude phase inconsistency caused by topography, and on this basis, proposed an InSAR registration algorithm based on local topography, and experimented for the algorithm with the same data, experimental process and results to verify the algorithm, and the registration results of the other two methods were also compared and analyzed.2) Optimal selection of sequence SAR interference pair. The selection of the SAR interference pair can affect the coherence and coherent point target extraction. Starting from the principle of InSAR Coherence Estimation to summarize the InSAR decorrelaiton factors, and then according to the time baseline, spatial baseline and Doppler centroid frequency of the three factors to do experiments on public master image selection and sequence of SAR images pairs selection based on minimum spanning tree. Then, from a methodological principle and the experimental results, the characteristics and application range of these two methods were compared and summarized. In this basis, all the coherent assessment values are projected to the three-dimensional space, then an BOX selection algorithm which based on cluster theory of interference pairs are proposed. Further treatment for the sequence characteristics of SAR differential interferometry to study the time continuity requirements of the sequence SAR differential interferometric processing, and the time baseline factor is transformed into continuity in time condition. Finally, an time continuity enhanced Box selection algorithm is proposed and experiments of this algorithm are also carried out. The algorithm satisfies the coherence and time continuity requirements.3) Application on using D-InSAR technique for deformation detecting of different types of geological disasters. The goal of this dissertation is how to improve detection capacity of the D-InSAR technique on deformation of geological disasters, to this end, two different types of geological disasters deformation detection application are applied. One is to choose Changzhou city as a strong coherent area for urban settlement deformation detection, and the other is to choose XinTan at Three Gorges as a weak coherent area for landslide deformation detection. Application effects on different sequence differential SAR interferometry methods based on coherent point target are studied in this dissertation, through the analysis of deformation detection results, the effectiveness of different sequence differential SAR interferometry methods based on coherent point target applied to different types of disasters deformation detection were evaluated.4) Development of InSAR data processing and simulation prototype software system. According to dissertation research and application needs, InSAR data processing and simulation software is studied and developed. Software architecture, plug-in mechanism architecture, base classes design as well as development and operation environment are described in this dissertation, and functional module design and implementation are introduced in detail. This software has implemented the whole InSAR process flow from data input, processing to the output of processing results and also simulation of SAR and InSAR, some preliminary processing results are given in the final.Finally, the study contents and innovations are summarized, and future research directions and research ideas were discussed as well.