Abstract
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). Two procedures for selecting the characteristics of the regions are then discussed. 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. It is also shown that dual polarization SAR data can yield segmentation resultS similar to those obtained with fully polarimetric SAR data.
Highlights
THE electromagnetic wave transmitted by a radar system is characterized by its frequency and its polarization state which describes the relative motion of the vector representing the electrical field when the wave moves towards an observer
Section VI1 gives several examples using real multilook polarimetric synthetic aperture radar (SAR) complex data to illustrate the potential of the selection procedures and evaluate the performance of the MAP technique
The MAP method produces better looking and more homogeneous regions than the maximum likelihood (ML) method, yet, as opposed to what would be expected with a straightforward box filtering technique, detection accuracy is not impaired as the metallic trihedral comer reflectors and other small features about a few pixels wide are still present after optimization of the region labels
Summary
THE electromagnetic wave transmitted by a radar system is characterized by its frequency and its polarization state which describes the relative motion of the vector representing the electrical field when the wave moves towards an observer. There is a considerable interest in developing segmentation procedures for polarimetric SAR data as they may yield a more quantitative, time, and cost effective analysis of the data than visual interpretation of photo products They may facilitate the inference of geophysical parameters from the surface and the near surface (e.g., surface roughness, soil moisture content, dielectric constant, etc.) as model inversion techniques are computationally more efficient when applied on the polarimetric backscatter characteristics of large homogeneous regions than on a pixel by pixel basis [8]. Section VI1 gives several examples using real multilook polarimetric SAR complex data to illustrate the potential of the selection procedures and evaluate the performance of the MAP technique.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
