Baseline Estimation And Geocoding Of Spaceborne Insar Techniques In Difficult Mapping Areas

Synthetic Aperture Radar Interferometry (InSAR) technology can realize all-time, all-weather observation on earth, and extract large-scale, high-precision three-dimensional surface information fast. High-resolution spaceborne InSAR technology provides possibilities to extract digital elevation model (DEM) from Difficult Mapping Areas such as Hengduan Mountains. In this thesis, aiming at the characteristics of large undulant terrain in Difficult Mapping Areas and the utility of InSAR processing software, the key processing steps are studied, including baseline estimation, height computation and geocoding. All those provide the technical foundation and technical support for InSAR processing software of Chinese independent property rights.The main work and innovations of this thesis are list as follows:(1) As a single algorithm can not meet the needs of practical application, this thesis adopts multi-level baseline estimation strategy from rough to precise estimation. In that method Ground Control Points(GCPs) are used to estimate accurate baseline based on rough baseline estimation. The difficult areas existed large undulant terrain and need high precision of baseline estimation. In order to reduce the model error, this thesis presents a baseline estimation method of integrating ellipsoid and GCPs combined with stochastic models and associated estimation algorithms (joint adjustment).(2) Common height computation algorithms are non-consistent in accuracy and efficiency. This paper presents an improved height computation algorithm, which improves the efficiency significantly and ensures the accuracy, by drawing on the three-dimensional expression of the baseline vector, using vector decomposition to solve look vector and avoiding the iterative solution.(3) It is difficult to obtain GCPs for Difficult Mapping Areas. Therefore this thesis presents a DEM geocoding algorithm based on sparse GCPs, least-squares are used to modify values by indirect geocoding, and the method is proved effectively to improve geocoding accuracy.(4) Combined with the western mapping project, using COSMO and TerraSAR high-resolution SAR images, the applicability of the algorithm proposed in this paper is proved. (5) Based on Visual C++ 6.0 platform, the baseline estimation, height computation and geocoding modules are developed as the important components of InSAR module and embed the SAR Mapping workstation software.