Region-Based Sea Ice Mapping Using Compact Polarimetric Synthetic Aperture Radar Imagery with Learned Features and Contextual Information

Operational sea ice maps are usually generated manually using dual-polarization (DP) synthetic aperture radar (SAR) satellite imagery, but there is strong interest in automating this process. Recently launched satellites offer compact polarimetry (CP) imagery that provides more comprehensive polarimetric information compared to DP, which compels the use of CP for automated classification SAR imagery. Existing scene algorithms rely on handcrafted features, while neural networks potential features discriminating. We have developed a new and effective algorithm leverages nature data. First, residual-based convolutional network (ResCNN) implemented classify each pixel. In parallel, an unsupervised segmentation performed generate regions based statistical properties. Regions assigned single class label by majority voting ResCNN output. For testing, quad-polarimetric (QP) scenes from RADARSAT Constellation Mission (RCM) used, QP, CP, reconstructed QP modes accuracy, also comparing them other approaches. Using achieves overall accuracy 96.86%, comparable (97.16%), higher than DP data about 2% 10%, respectively. The improved option mapping.