PolSAR Image Classification

PolSAR image classification aims to automatically categorize land cover types from polarimetric synthetic aperture radar data, a challenging task due to data complexity and limited labeled samples. Recent research focuses on leveraging deep learning, particularly convolutional neural networks (CNNs) and self-supervised learning techniques like contrastive learning, often incorporating novel architectures such as dual-branch or heterogeneous networks to better exploit multi-feature information and address the scattering confusion problem. These advancements improve classification accuracy, especially in scenarios with limited labeled data, enabling more efficient and accurate land cover mapping for applications in environmental monitoring, agriculture, and disaster response.