Investigation Of Vegetation Height Retrieval Of Polarimetric Interferometry SAR

Polarimetric interferometry SAR is a new technique that integrates polarization SAR and interferometry SAR. This new technique not only reflects the physical characteristics but also obtains spatial distribution properties of targets. It separates the interferometric phase center of multiple scattering mechanisms by scattering matrix decomposition in a resolution cell. Then polarimetric interferometry SAR provides the possibility for measurement of the terrain under vegetation and vegetation height. Thus, polarimetric interferometry SAR can apply to vegetation parameters inversion and the ground resources and environment monitoring.This paper studies the basic theories of polarimetric interferometry SAR and inversion algorithms of ground vegetation height. The main work and innovation of this paper are as follows:(1)Polarimetric interferometric SAR scattering phase center estimation algorithms are researched. The scattering phase center estimation algorithms are important to vegetation height inversion, thus this dissertation intensively studies the scattering phase estimation based on ESPRIT algorithm and Freeman polarimetric interferometry covariance matrix decomposition algorithm; Finally, the simulated data and real data are used to validate these algorithms. The results demonstrate that the scattering phase estimation based on ESPRIT algorithm can estimate the phase corresponding to the top of the canopy, but can not accurately estimate the phase corresponding to the ground. However, the scattering phase estimation based on Freeman-polarimetric interferometry covariance matrix decomposition algorithm can effectively estimate the ground phase.(2)By combing the advantages of the scattering phase estimation based on ESPRIT algorithm and that of Freeman polarimetric interferometry covariance matrix decomposition algorithm, a modified algorithm based on ESPRIT is proposed. It can solve the problem inaccurate ground phase of traditional ESPRIT algorithm. Finally, the experiments of the simulated data and real data demonstrate that the modified ESPRIT algorithm can obtain more accurate result.(3)The dissertation proposes a inversion algorithm based on the fusion of RVOG model and Freeman-polarimetric interferometry covariance matrix decomposition algorithm. It can estimate vegetation height by taking the ground phase of Freeman-polarimetric interferometry covariance matrix decomposition algorithm as the initial ground phase of the RVOG model. Finally, the experiment demonstrates that the proposed algorithm can obtain the best result. Figures:55, Tables:6, References:52.