LAUSR

At present, how to achieve high-precision hyperspectral image classification (HSIC) under the condition of few samples is a hot research issue. Metric-based meta-learning methods have proved to be very successful in this field. However, in terms of quantifying the dependencies between embedded features of hyperspectral samples, previous methods either only model marginal distribution and ignore joint distribution, limiting the expressive capability of feature representation, or bring large computational costs though considering joint distribution. In this letter, we propose a novel few-shot learning (FSL) method based on Brownian distance covariance (BDC) for HSIC, which learns hyperspectral images’ (HSIs) representations by measuring the discrepancy between joint characteristic functions of embedded features and product of the marginals. In addition, a lightweight feature extraction network based on tied block convolution is proposed to better model cross-channel correlation and aggregate global spectral–spatial features across channels. Extensive evaluations of several datasets show the effectiveness of the proposed method.