Research On High Precision Elevation Inversion With Multi-baseline

Interferometric Synthetic Aperture Radar Measurement Technology(InSAR),which can obtain elevation information of scene targets,has become one of the important ways of terrain altimetry.The airborne and spaceborne platforms are currently commonly used InSAR mapping platforms,and the onboard platform imaging and inversion is more complex than the airborne platform.This thesis is aimed at multi-baseline InSAR elevation inversion technology for airborne platforms and spaceborne platforms.Highprecision imaging algorithms,unwrapping algorithms,geometric correction and elevation inversion algorithms are studied to obtain elevation information of scene targets.The specific work and innovations of this thesis are as follows:(1)A multi-baseline optimization weighted least squares phase unwrapping algorithm based on the quality map is proposed.Based on the multi-baseline least squares algorithm,the algorithm compensates the residues by using the quality map as the prior information,and sets the threshold to optimize the unwrapping algorithm of low-quality region by using the weighted second-order phase gradient.The weighted algorithm not only overcomes the layover effect in the single-baseline unwrapping algorithm,but also improves the unwrapping accuracy of the proposed algorithm compared with the existing multi-baseline least squares algorithm.Results of simulation and measurement data show the effectiveness and the improved performances of the proposed algorithm.(2)A progressive high-precision elevation inversion scheme combining the back projection of the surface with multiple baselines is constructed,and a terrain geometric distortion correction method is optimized.The surface projection algorithm has the advantages of integration of imaging registration and de-leveling effect,and the flexible selection of reference planes is easy to phase unwrapping.This thesis combines the surface projection algorithm with multi-baseline data to construct a progressive elevation inversion method.Firstly,the elevation obtained by the short baseline data is used as the initial reference surface,then the baseline data of different lengths are used to perform surface imaging and interference processing,and the higher-precision elevation reconstruction is successively obtained.The inversion accuracy is verified by multibaseline InSAR simulation.In addition,the radar squint will cause the tilting offset of the inversion terrain.Aiming at the zero error value in the original scene after terrain correction,a method similar to noise detection and median filtering is used to compensate the error.Finally,more effective and more accurate inversion terrain can be obtained.(3)Provides a high-precision elevation inversion method for overcoming the spherical effect under the spaceborne wide swath.In view of the influence of the curvature of the Earth's spherical surface during the imaging of the large-spot area,the target scene is imaged and elevation-inferred according to the latitude and longitude of the Earth's latitude and longitude,which effectively solves the image quality caused by the curvature of the Earth during the spaceborne wide swath.At the same time,the influence of the rotation speed of the Earth is also considered in the imaging process,and the prior art DEM information is added to the scene of the surface elevation change for surface projection imaging and interference processing.The simulation results show that compared with the traditional uniform planar meshing and projection,the method provided in this thesis improves the quality of on-board imaging and obtains high elevation inversion accuracy.