Dual-Branch Residual Network for Cross-Domain Few-Shot Hyperspectral Image Classification with Refined Prototype

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 2025 1 Dual-Branch Residual Network for Cross-Domain Few-Shot Hyperspectral Image Classification with Refined Prototype Anyong Qin, Chaoqi Y uan, Qiang Li, Feng Y ang, Tiecheng Song and Chenqiang Gao Abstract --Convolutional neural networks (CNNs) are effective for hyperspectral image (HSI) classification, but their 3D convolu-tional structures introduce high computational costs and limited generalization in few-shot scenarios. Domain shifts caused by sensor differences and environmental variations further hinder cross-dataset adaptability. Metric-based few-shot learning (FSL) prototype networks mitigate this problem, yet their performance is sensitive to prototype quality, especially with limited samples. T o overcome these challenges, a dual-branch residual network that integrates spatial and spectral features via parallel branches is proposed in this letter . Additionally, more robust refined prototypes are obtained through a regulation term. Experiments on four publicly available HSI datasets illustrate that the proposal achieves superior performance compared to other methods.