Research Of Spaceborne SAR Image Positioning And Application Of Rational Function Model In InSAR

The state of art of Spaceborne SAR system is rising rapidly, the spatial and temporalresolutions of which are higher and higher. SAR system plays a more important role in the fieldof earth observation. It’s pressing to carry out more research on spaceborne sar image. Oneimportant application of Spaceborne SAR image is positioning with sparse(no) ground controlpoints(GCPs), which is capable of achieving the cartesian coordinate of crucial target insub-controled areas. Rational Function Model(RFM) is widely used in data processing of opticaland SAR photogrammetry. It’s of great significance both theoretically and practically tointroduce RFM in InSAR data processing. The thesis took the application of High resolutionspaceborne SAR images as its thread, and analyzed the rigorous and universal geometric modelof SAR image, respectively researched in target positioning and the application of RFM model inInSAR data processing. The main work done in the thesis is listed as follows.1) The significance and development status of SAR and InSAR technology is summarizedand analyzed in the aspect of loading instrument, imaging characteristic and data processing,thus the main contents of research is acertained.2) Principle of target positioning without ground control information based on…Range-Doppler model is illustrated and analyzed. Experiment using TerraSAR Spotlight imagewith a spatial resolution of1m is done without GCPs. The result shows the positioning accuracyare respectively50m and70m in range direction and azimuth direction. To improve the accuracy,a small amount of GCPs is introduced to refine the positioning parameters of R-D model, thisstep improved the accuracy to20m and25m in range direction and azimuth direction. Then anew approach of conversing target height to the equator radius is presented, and a quantitativeanalysis of its effect on positioning accuracy is done based on deducrtion and simulation.3) The rigourous sensor model of InSAR is eatablished based on InSAR geometric…principle. The similarity and difference between InSAR and optical photogrammetry is analyzedto declare the feasibility of adapting RFM in InSAR data processing.4) A group of InSAR-RFM equations are established and used for image registering…flatting-phase eliminating and phase-height converting. An algorithm to resolve RFM coefficentswithout iteration and initial value is presented based on global low resolution DEM. Experimentis done on SRTM simulated data and simulated data in Tian’Shan areas, the replacementaccuracy of RFM reaches0.1941m in SRTM data and0.000014m in Tian’Shan data. To solvethe ill-conditioned problem of resolving RFM coefficents, Direct computing method and errorcompensating method are respectively applied, and the result shows the accuracy had risen 17.9%with direct method and44.8%with error compensation method in SRTM simulated data.