Water extraction in SAR images using GLCM and Support Vector Machine

Traditional methods to extract water regions in SAR images usually rely on image binarization with a specified threshold. However, because of the inherent speckles in SAR images, finding an appropriate threshold is very difficult. In the paper, we propose a new method for water region extraction in SAR images using GLCM (gray-level co-occurrence matrix) based features combined with SVM (Supported vector Machine). The characteristics of water and non-water regions are distinctively depicted by GLCM based features, which are fed into the SVM classifier to extract water regions. Experiments on synthetic and real SAR images demonstrate that the proposed method achieves better results compared with two other ones.