Support vector machine classification of land cover: application to polarimetric SAR data

This paper addresses a novel SVM-based classification scheme of land cover from polarimetric synthetic aperture radar (SAR) data and discusses some important properties of SVMs, for example the relation between the number of support vectors and classification accuracy.

Abstract

Support vector machines (SVMs) have much attention as a promising approach to pattern recognition. They are able to handle linearly nonseparable problems 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. The SVMs are successfully applied to the feature vectors which consist of several polarimetric features or the texture measure, and perform efficient image classification. Some important properties of SVMs, for example the relation between the number of support vectors and classification accuracy, are also discussed.