EUVIMEDEUVIMED entdecken
Crossref

SoDa 2 : Single-Stage Open-Set Domain Adaptation via Decoupled Alignment for Cross-Scene Hyperspectral Image Classification

Yiwen Liu, Minghua Wang, Jing Yao, Xin Zhao, Gemine Vivone

IEEE Transactions on Geoscience and Remote Sensing · 2026 · 64 · 4705513-4705513

DOI: 10.1109/tgrs.2026.3707690 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2026. Verfasst von Yiwen Liu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 45 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Tianjin, Fundamental Research Funds for the Central Universities.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cross-scene hyperspectral image (HSI) classification stands as a fundamental research topic in remote sensing, with extensive applications spanning various fields. Owing to the inclusion of unknown categories in the target domain and the existence of domain shift across different scenes, open-set domain adaptation (OSDA) techniques are commonly employed to address cross-scene HSI classification. However, existing OS cross-scene HSI classification methods still face two critical challenges: 1) domain shift issues arising from the direct alignment of mixed spectral–spatial features and 2) high computational costs caused by two-stage training strategies. To address these issues, this article proposes a single-stage open-set domain adaptation method based on decoupled alignment (SoDa2) for cross-scene HSI classification. A contribution-aware dual-modality feature extraction is customized to disentangle the characteristics from spectral sequence signals and spatial details, selectively and adaptively enhancing discriminative features. The decoupled alignment module minimizes the maximum mean discrepancy (MMD) to independently reduce the spectral discrepancy and the spatial discrepancy between the source and target domains, extracting more fine-grained domain-invariant features. A cost-effective single-stage dual-branch framework is designed to learn MMD-constrained aligned features and constraint-free intrinsic features for adaptive distinction between known and unknown classes. This framework employs a Gaussian mixture model (GMM) to model the squared cosine similarity distribution between the two feature types, enabling OS recognition without prior knowledge of unknown classes. Extensive experiments on three groups of HSI datasets demonstrate that SoDa2 outperforms state-of-the-art methods, achieving superior classification accuracy and model transferability for OS cross-scene tasks. The SoDa2 code will be available at https://github.com/liuyiwen523/SoDa2

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Tianjin
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Liu, Y., Wang, M., Yao, J., Zhao, X., & Vivone, G. (2026). SoDa 2 : Single-Stage Open-Set Domain Adaptation via Decoupled Alignment for Cross-Scene Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 64, 4705513-4705513. https://doi.org/10.1109/tgrs.2026.3707690

Vancouver

Liu Y, Wang M, Yao J, Zhao X, Vivone G. SoDa 2 : Single-Stage Open-Set Domain Adaptation via Decoupled Alignment for Cross-Scene Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing. 2026;64:4705513-4705513. doi:10.1109/tgrs.2026.3707690

BibTeX
@article{Liu2026,
  author = {Liu, Yiwen and Wang, Minghua and Yao, Jing and Zhao, Xin and Vivone, Gemine},
  title = {SoDa 2 : Single-Stage Open-Set Domain Adaptation via Decoupled Alignment for Cross-Scene Hyperspectral Image Classification},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2026},
  volume = {64},
  pages = {4705513--4705513},
  doi = {10.1109/tgrs.2026.3707690}
}
Kostenlos starten →