LAUSR

Nowadays, domain adaptation (DA) is getting more attention in cross-scene hyperspectral image classification (HSIC), and various DA algorithms have been proposed. However, regular convolution indiscriminately extracting features around the center pixel will result in the inaccurate extraction of spatial–spectral features, which significantly affects the subsequent feature alignment (FA). Meanwhile, the method of aligning the category features of source and target domains (TDs) from a single level may not cope well with complex HSIs. Therefore, we propose a multilevel FA algorithm based on spatial attention deformable convolution (MFA-SADC), which achieves MFA from feature to feature (F-to-F), feature to cluster-center (F-to-C), and cluster-center to cluster-center (C-to-C). In addition, spatial attention deformable convolution (SAD-Conv) is proposed to compose the feature extraction network of MFA-SADC, which guarantees the purity of spatial–spectral features. Experiments on three HSI datasets indicate that MFA-SADC can obtain better classification performance when compared with the seven state-of-the-art methods.