Optimization Research On Polarimetric Calibration Methods Of Airborne SAR Based On Point Target

With the development of Polarimetric Synthetic Aperture Radar(PolSAR)technology,polarimetric SAR data has been widely applied to target detection,earth surface parameter inversion and ground objects classification,and polarimetric SAR calibration technology is the premise for usage of polarimetric SAR data in various fields.The key to polarimetric SAR calibration is to eliminate the influence of polarimetric distortion caused by the incomplete isolation of electromagnetic waves between channels,so that the amplitude and phase under different polarimetric states can accurately reflect the ground objects information for polarimetric SAR data.For the polarimetric calibration technology,a lot of research work has been done by the predecessors.At present,the polarimetric calibration methods can be roughly divided into three categories,including polarimetric calibration technique based on point targets,distributed targets and hybrid targets.In the process of studying the algorithms based on point target polarimetric calibration technology,it is found that the unconstrained algorithm fails to fully utilize the feature target,resulting in a certain precision error of the obtained parameter matrix.Aiming at this problem,this paper proposes an improved polarimetric calibration algorithm based on unconstrained algorithm.On the basis of the original model and hypothesis,Newton iterative algorithm with least squares principle is introduced to optimize the calibration process for calibration parameters solution and to improve algorithm accuracy.Among them are the following research and innovation work:(1)The basic theory of polarimetric SAR calibration is introduced.The content explains the polarimetric representation of electromagnetic waves,the polarimetric scattering characteristics of target objects,the polarimetric scattering matrix,the polarimetric synthesis of images,and the polarimetric characteristic response of point targets,the causes of polarimetric distortion,radiation calibration and the main factors affecting the accuracy of polarimetric calibration in detail.The accuracy evaluation method of polarimetric calibration method based on point target is described in detail.(2)The basic model of airborne SAR polarimetric calibration is described,and the Whitt algorithm and the unconstrained algorithm are derived based on this model.Aiming at the problem of the various assumptions in the calibration methods will reduce the accuracy of polarimetric calibration,an improved unconstrained algorithm is proposed,the algorithm is based on the redundant observations of the calibration data.The least squares principle and Newton iteration are introduced in the process of solving the parameters of the polarimetric calibration matrix.The method can improve the accuracy of the polarimetric SAR image calibration algorithm to some extent.(3)The data of the Danling,Sichuan area obtained by the airborne X-band full-polarimetric SAR system was used to verify the Whitt algorithm,unconstrained algorithm and improved algorithm.The experimental results can be verified in the following different ways:1)Perform Pauli decomposition on the image before and after calibration,and verify the consistency of the color mechanism and scattering mechanism of the image after decomposition;2)Verification by polarization signature;3)verification by amplitude error;4)verification by phase error.The results show that compared with the unconstrained algorithm and the Whitt algorithm,the proposed method can slightly improve the polarimetric calibration precision.