Extraction of Antarctic Ice Features Using Hybrid Polarimetric RISAT-1 SAR Data

Compact polarimetry has gained popularity due to its advantages, such as larger swath, simple architecture and low power consumption. The backscattered signal and scattering decomposition vary for different targets based on their electrical, geometrical and structural properties. As of now, the potential of hybrid polarimetric synthetic aperture radar (SAR) data for exploring Antarctic ice features is not fully explored. Here, we present a comprehensive polarimetric feature analysis and classification results of the hybrid polarimetric dataset acquired by RISAT-1 near the Indian Antarctic research station Maitri. The single-look complex images have been subjected to polarimetric data processing for extracting Antarctic ice features using POLSARPRO software. The polarimetric coherence matrix is generated and then filtered to eliminate speckles. Raney m–χ decomposition technique has been utilized to understand the scattering mechanism of the targets. The decomposed RGB image is classified using Wishart-supervised classification, and classification accuracy is assessed using a confusion matrix. It is found that the comparatively simple hybrid polarimetric SAR provides sufficient information to detect and discriminate various Antarctic ice features. Features such as rifts, ice–rises, ice shelves and icebergs are clearly discriminated using Wishart-supervised classification. It is also found that the overall accuracy of the classification of study areas varies between 80% and 97%, suggesting a good classification outcome.