Speckle statistically brings series connections of dark pixels, which can be observed as dark line features in synthetic aperture radar (SAR) images. The dark lines have no physical meaning. In this paper, line features of that kind in high-resolution SAR images whose intensity obeys a K-distribution are studied. It is stochastically explained that the dark line features in 1-look K-distributed images can be observed more distinctly than those in exponential distributed images. It is further revealed that such line features are detectable enough, even if the K-distributed images are multilooked. The experiments on simulated images as well as on actual SAR images confirm the explanation.

Support vector machines (SVMs), newly introduced in the 1990s, are promising approach to pattern recognition. They are able to handle linearly nonseparable problems without difficulty, by combining the maximal margin strategy with the kernel method. This paper addresses a novel SVM-based classification scheme of land cover from polarimetric synthetic aperture radar (SAR) data. Polarimetric observations can reveal existing different scattering mechanisms. As the input into SVMs, the polarimetric feature vectors, composed of intensity of each channel, sometimes complex correlation coefficients and textural information, are prepared. Classification experiments with real polarimetric SAR images are satisfactory. Some important properties of SVMs, for example the relation between the number of support vectors and classification accuracy, are also investigated.