Cross-calibration Method Of Polsar Co-polarization Channel Imbalance Based On Stable Targets

Polarimetric Synthetic Aperture Radar(Pol SAR)is widely used in the fields of land cover classification,surface parameter inversion,and target identification.Errors such as crosstalk,cross-polarization channel imbalance,and co-polarization channel imbalance in Pol SAR observation data will adversely affect the performance of various applications,and errors need to be corrected by polarimetric calibration before application.Traditional polarimetric calibration methods are often based on manual calibrators when estimating the co-polarization channel imbalance;however,due to the limitation of the number of manual calibrators and the difficulties of field deployment,it is hard to improve the frequency of polarization calibration.Methods based on natural distribution targets to estimate the copolarization channel imbalance have been rapidly developed in recent years,but the existing methods often require the distribution targets as the true value reference to meet specific constraints.In this thesis,the idea of crosscalibration was introduced,which uses the calibrated Pol SAR to obtain the true value of polarimetric scattering of arbitrary distributed targets so as to reduce the constraints on the targets of traditional methods.In this way,it is also possible to improve the frequency of calibration.This thesis focuses on solving the problems of accurate selection of stable targets and effective utilization of statistical information about stable targets in cross-calibration of fully Pol SAR with co-polarization channel imbalance and further extends the method to the study of dual Pol SAR calibration with less polarimetric scattering information.The main work and innovations are as follows:(1)A stable target extraction method based on the cross-polarization scattering ratio parameter is studied.Cross-calibration of Pol SAR requires that distribution targets stay the same over time,and the co-polarization imbalance error makes it hard to find stable targets in uncalibrated data.Since the ratio of different polarization channels can cancel out the copolarization imbalance error,the cross-polarization scattering ratio parameter is created by combining the ratio of different polarization channels.It has been verified through experiments that the proposed method is not easily affected by the size of the co-polarization channel imbalance error and can extract stable targets for a variety of scattering types.(2)A cross-calibration method based on the statistical properties of stable targets with the co-polarization channel imbalance is studied.The method used the true polarimetric scattering value of the stable targets in the calibrated data to cross-estimate the co-polarization channel imbalance of the stable targets in the uncalibrated data and obtained reasonable statistical methods for estimating the value through theoretical and experimental analysis to overcome the influence of the abnormal estimation on the accuracy of calibration.Finally,considering that the co-polarization channel imbalance pair will change with the radar incidence angle,the final copolarization channel imbalance estimation is obtained by the least squares method.It is experimentally verified that the minimum error of the method is 5.8° for phase estimation and 0.24 d B for amplitude estimation.(3)The cross-calibration model of co-polarization channel imbalance for dual Pol SAR is studied.In order to solve the problem that there is very little polarization information in dual Pol SAR data and only a small amount of phase information that can be extracted,it is made to extract the phase stability targets by using the strong polarimetric scattering ratio,and the results of co-polarization channel phase and amplitude estimation are counted by the median method and the mean method,respectively.It is experimentally verified that the method has an error of 2.5° in phase estimation and a minimum error of 0.007 d B in amplitude estimation,with an average error of 0.3 dB.