EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LGDAF-Net: A Lightweight CNN–Transformer Framework for Cross-Domain Few-Shot Hyperspectral Image Classification

Guang Yang, Jiaoli Fang, Daming Zhu, Xiaoqing Zuo

Electronics · 12.04.2026 · 15 (8) · 1606

DOI: 10.3390/electronics15081606 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 12.04.2026. Verfasst von Guang Yang und 3 weiteren Autor:innen. Die Arbeit nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Yunnan International Joint Laboratory for Integrated Sky-Ground Intelligent Monitoring of Mountain Hazards, Yunnan Provincial Key Laboratory for Intelligent Monitoring of Natural Resources and Spatio-temporal Big Data Governance.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cross-domain few-shot hyperspectral image (HSI) classification is challenging due to limited labeled samples and distribution shifts across sensors and acquisition scenes, which often degrade feature representation and classification performance. This study proposes a lightweight hierarchical CNN–Transformer framework, termed LGDAF-Net (Lightweight Global and Local Dual Attention Fusion Network), for effective cross-domain few-shot HSI classification. The framework progressively enhances spectral–spatial representation through three stages: spectral–spatial feature recalibration, local spatial structure perception, and global contextual modeling. Specifically, a spectral–spatial dual-attention enhancement module (SESA) is introduced to emphasize informative spectral responses and suppress redundancy. A Local Attention Spatial Perception Module (LASPM) is designed to capture fine-grained spatial structures, while a lightweight Transformer-based Global Attention Context Modeling Module (GACM) models long-range spatial dependencies. In addition, kernel triplet loss and domain adversarial learning are incorporated to improve feature discrimination and promote cross-domain feature alignment. Experimental results on three benchmark datasets demonstrate that the proposed method achieves competitive performance compared with existing methods.

Förderung

  • Yunnan International Joint Laboratory for Integrated Sky-Ground Intelligent Monitoring of Mountain Hazards
  • Yunnan Provincial Key Laboratory for Intelligent Monitoring of Natural Resources and Spatio-temporal Big Data Governance

Zitieren

APA

Yang, G., Fang, J., Zhu, D., & Zuo, X. (2026). LGDAF-Net: A Lightweight CNN–Transformer Framework for Cross-Domain Few-Shot Hyperspectral Image Classification. Electronics, 15(8), 1606. https://doi.org/10.3390/electronics15081606

Vancouver

Yang G, Fang J, Zhu D, Zuo X. LGDAF-Net: A Lightweight CNN–Transformer Framework for Cross-Domain Few-Shot Hyperspectral Image Classification. Electronics. 2026;15(8):1606. doi:10.3390/electronics15081606

BibTeX
@article{Yang2026,
  author = {Yang, Guang and Fang, Jiaoli and Zhu, Daming and Zuo, Xiaoqing},
  title = {LGDAF-Net: A Lightweight CNN–Transformer Framework for Cross-Domain Few-Shot Hyperspectral Image Classification},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {8},
  pages = {1606},
  doi = {10.3390/electronics15081606}
}
Kostenlos starten →