Polarimetric SAR image classification using binary coding‐based polarimetric‐morphological features

Polarimetric synthetic aperture radar (POLSAR) systems provide high resolution images containing polarimetric information. So, they have capability in land cover classification. In this work, a binary coding-based polarimetric-morphological (BCPM) feature extraction is proposed for POLSAR image At first, set of features proposed. Then, new morphological framework introduced contextual from the cube. The coherence matrix composed diagonal and non-diagonal elements with different These are analysed separately method. Moreover, amplitude phase components individually using filters by reconstruction. Finally, polarimetric-spatial reduction, which uses first order statistics, transformation. experiments on three real dataset show superior performance BCPM compared to several classification methods.