Research On The Methods In Processing Of Fully Polarimetric Synthetic Aperture Radar Imagery

Polarimetric synthetic aperture radar imaging is a technique based on microwave remote sensing. Polarimetric synthetic aperture radar can provide large-scale two-dimensional high spatial resolution images of the observed objects. In addition, with the wide availability of PolSAR data from both space borne and airborne SAR systems, the research on processing of PolSAR data is developing fastly. For the past recent years, significant advances in polarimetric synthetic aperture radar instruments and information extraction techniques have flourished, and related research and applications have reached a certain degree of maturity. However, part of the techniques in this field deserves a thorough investigation. Synthetic aperture radar systems come along from single polarization and monochromatic imaging to multi-polarization and multi-frequency imaging. The later is also called the multi-channel radar imagery. Using multi-channel radar system for observing the earth can acquire much more information than single channel radar system. In this thesis, three problems, including speckle filtering, bias correction for eigen-decomposition parameters and unsupervised classification, in the processing of monostatic and monochromatic PolSAR data will be studied.At first, the description method for polarization state of electromagnetic wave is presented. Then the geometrical configuration and principles of imaging system in polarimetric synthetic aperture radar is put forward. The open-source platform by the European space agency and mainstream airborne datasets are listed.Thereafter a simple analysis is put forward at first about the principles in speckle filtering. One of the most important clauses is preserving the polarimetric properties and reducing the speckle level in the data. The coherency matrix and covariance matrix after speckle filtering must be semi-definite positive. A study on the behavior and performance of the nonlinear anisotropic diffusion equation is accomplished. Whereafter an extension to the multi-dimensional nonlinear anisotropic diffusion is done. The total scattered power is employed to construct the diffusivity function and the edge-enhancing diffusion scheme is adopted to design the diffusion tensor. At last the algorithm of speckle filtering using nonlinear anisotropic diffusion is listed. It can be easily observed from the experimental results that the new proposed method can preserve the polarimetric properties and reduce the speckle level.Whereafter in the calculation of eigen-parameters for polarimetric target decomposition, the ensemble averaging is substituted by the spatial averaging. Thus the estimation for coherency matrix or covariance matrix will contain bias, which will be studied in this thesis. Consequently the statistics of these parameters are analyzed at first. The simulated data generated by the PolSARpro platform is used for the reason that it has dodged the influence of the artifact and system noise. The simulated data can be viewed as a realized data. An asymptotically statistical analysis is performed on simulated data and thereafter the statistical behavior of averaged alpha angle and entropy are acquired. Based on the statistical behavior, it can be concluded that the influence caused by the window size on averaged alpha angle can be ignored. Nevertheless the entropy is sensitive to the windows size. After performing an averaging of many random points in simulated data, it can be seen that approximate linear relation between entropy and window size is valid. Bias correction algorithm is constructed based on the above approximate linear relation. It can be concluded from the experimental results that the entropy which has been corrected will enhance the accuracy of unsupervised classification using eigen-decomposition.At last, polarimetric classification is one of the most important applications in PolSAR imaging. The kernel idea of supervised classification is employed in this thesis. The scattering model based decomposition is combined with the supervised classification. The four-component scattering model based decomposition is used as the initial input of classification. The Wishart classifier is used iteratively. It is shown from the experimental results that the new unsupervised classification method possesses higher accuracy compared with the classic three-component method. Compare with the classic method based on eigen-decomposition, the new method stay away from the problem of uncertain upper-bound and lower-bound.