Adversarial Dilated Large-Kernel Attention Networks for Cross-Domain Few-Shot Hyperspectral Image Classification

The labeling hyperspectral image (HSI) data are time-consuming and laborious, resulting in limited labeled samples for training deep learning-based classifiers. To address this challenge, we propose ADLKAN, a cross-domain few-shot learning (FSL) framework based on an adversarial dilated large-kernel attention network. First, the ADLKAN consists of an adversarial dilated LKA feature extractor, which can enhance the ability to capture sparse features. The feature extractor includes two cooperative spectral–spatial attention blocks with large-kernel convolution, each of which extracts spectral and spatial features separately. These features are then fused to capture local contextual information and long-range dependencies in the spatial and spectral domains with attention networks. Second, ADLKAN also designs a multiscale conditional domain discriminator to alleviate the weakness of FSL under domain shift. The detailed multiscale features are extracted by three subdiscriminators on different input scales. Finally, the adversarial training is then adopted to optimize the embedding feature extractor and strengthen its consistency in uncertain regions of samples from different domains. Experimental results on five public HSI datasets show that the proposed ADLKAN outperforms other state-of-the-art methods.