Polarimetric synthetic aperture radar (PolSAR) has achieved a prominentposition as a remote imaging method. However, PolSAR images are contaminated byspeckle noise due to the coherent illumination employed during the dataacquisition. This noise provides a granular aspect to the image, making itsprocessing and analysis (such as in edge detection) hard tasks. This paperdiscusses seven methods for edge detection in multilook PolSAR images. In allmethods, the basic idea consists in detecting transition points in the finestpossible strip of data which spans two regions. The edge is contoured using thetransitions points and a B-spline curve. Four stochastic distances, twodifferences of entropies, and the maximum likelihood criterion were used underthe scaled complex Wishart distribution; the first six stem from the h-phiclass of measures. The performance of the discussed detection methods wasquantified and analyzed by the computational time and probability of correctedge detection, with respect to the number of looks, the backscatter matrix asa whole, the SPAN, the covariance an the spatial resolution. The detectionprocedures were applied to three real PolSAR images. Results provide evidencethat the methods based on the Bhattacharyya distance and the difference ofShannon entropies outperform the other techniques.
展开▼
