文献类型：期刊文献

英文题名：Cross-Scene Hyperspectral Image Classification via Bidirectional Mamba and Domain Mixing Network

作者：Dang, Junzhe[1,2,3] Guo, Chengwang[1,2,3] Zhang, Mengmeng[1,2,3] Zhang, Yuxiang[4] Jia, Wen[5,6] Li, Wei[1,2,3]

第一作者：Dang, Junzhe

通信作者：Zhang, MM[1];Zhang, MM[2]

机构：[1]Beijing Inst Technol, Sch Informat & Elect, Beijing 100081, Peoples R China;[2]Beijing Inst Technol, Natl Key Lab Sci & Technol Space Born Intelligent, Beijing 100081, Peoples R China;[3]Beijing Inst Technol, Zhuhai 519088, Guangdong, Peoples R China;[4]Univ Hong Kong, Dept Geog, Jockey Club STEM Lab Quantitat Remote Sensing, Hong Kong, Peoples R China;[5]Chinese Acad Forestry, Inst Forest Resource Informat Tech, Beijing 100091, Peoples R China;[6]Natl Forestry & Grassland Adm, Key Lab Forestry Remote Sensing & Informat Syst, Beijing 100091, Peoples R China

年份：2026

外文期刊名：IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS

收录：;EI(收录号：20260319922510);Scopus(收录号：2-s2.0-105027391845);WOS:【SCI-EXPANDED(收录号:WOS:001663482700001)】；

基金：This work was supported in part by China Postdoctoral Science Foundation under Grant 2025T181109, in part by the Postdoctoral Fellowship Program of China Postdoctoral Science Foundation (CPSF) under Grant GZC20252779, and in part by the National Natural Science Foundation of China under Grant 42101403.

语种：英文

外文关键词：Feature extraction; Training; Transformers; Computational modeling; Predictive models; Mathematical models; Hyperspectral imaging; Computer architecture; Forestry; Data models; Cross-scene; domain mixing; hyperspectral image (his) classification; mamba; unsupervised domain adaptation (DA)

摘要：To overcome the challenges posed by domain shift in hyperspectral image (HSI) classification, methods based on domain adaptation (DA) have been widely used. Currently, most HSI DA methods focus on designing complex strategies to align the distributions of the source domain (SD) and the target domain (TD) in the feature space after feature extraction, yielding promising results. However, when there exists a large domain shift between SD and TD, it becomes challenging to map them into the same feature space. In this article, we propose the bidirectional mamba and domain mixing network (BMDMnet). Since pure CNN architectures are constrained in local feature extraction, while transformer-based models improve global feature capturing capability at the cost of high computational complexity, we propose the bidirectional mamba module (BMM) as an efficient solution for capturing long-range dependencies. In addition, a self-distillation strategy is employed during training. By utilizing a more stable teacher model, reliable predictions can be obtained in the TD. Subsequently, a domain mixing supervised learning (DMSL) module is designed, which creates a mixed domain by selecting low-entropy sample-pseudo-label pairs from the TD and randomly combining them with sample-label pairs from the SD. DMSL aims to introduce mixed domain to mitigate the inter-domain gap in the data space, thereby enabling the model to learn TD representations more effectively. Experiments demonstrate that BMDMnet outperforms state-of-the-art algorithms across three cross-scene datasets.