Generation of Pseudo-fully Polarimetric Data for Land Cover

       2012年遥感计算器视觉国际会议——A linear relationship among the HH, HV, and VV components of polarimetric synthetic aperture radar (SAR) data is studied. A regression model was developed to predict the real and imaginary parts of the VV polarimetric component from the HH and HV components in dual polarimetric SAR and the resulting dataset is called pseudo-fully polarimetric SAR data. Freeman-Wishart classification was applied to evaluate the preservation of scattering characteristics in the pseudo-fully polarimetric dataset. A kappa coefficient is 0.81 indicates very good agreement between the two classification results. An SVM was used for the land cover classification. Finally, post-processing was implemented to remove noise in the form of isolated pixels. AVNIR-2 optical data taken over the same area at nearly same time was used as ground truth data to assess the classification proposed algorithm deals with two tasks: traffic signs detection and traffic signs recognition. Firstly, multi-scale phase spectrum of quaternion Fourier transformation method is used to obtain the location of traffic signs in scenes image. Secondly, traffic signs local sparse features are extracted by the improved algorithm based on SURF descriptors and locality-constrained linear coding (LLC) method. Finally, linear support vector machine (SVM) is used to train classifier and test recognition accuracy rate of ban traffic signs. Extensive experiments on 1000 images show that our approach can improve recognition accuracy rate and reduce running time.