Segmentation of polarimetric synthetic aperture radar data

A statistical image model is proposed for segmenting polarimetric synthetic aperture radar (SAR) data into regions of homogeneous and similar polarimetric backscatter characteristics. A model for the conditional distribution of the polarimetric complex data is combined with a Markov random field representation for the distribution of the region labels to obtain the posterior distribution. Optimal region labeling of the data is then defined as maximizing the posterior distribution of the region labels given the polarimetric SAR complex data (maximum a posteriori (MAP) estimate). An implementation of the MAP technique on a parallel optimization network is presented. A fast alternative solution is also considered. Two procedures for selecting the characteristics of the regions are then discussed: one is supervised and requires training areas, the other is unsupervised and is based on the multidimensional clustering of the logarithm of the parameters composing the polarimetric covariance matrix of the data. Results using real multilook polarimetric SAR complex data are given to illustrate the potential of the two selection procedures and evaluate the performance of the MAP segmentation technique. The impact of reducing the dimension of the polarimetric measurements on segmentation accuracy is also investigated. The results indicate that dual polarization SAR data may yield almost similar segmentation results as the fully polarimetric SAR data.