EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

FAD-Net: Frequency Alignment Dual-Branch Network for Hyperspectral Image Super-Resolution

Junge Bo, Chen Ling, Ji-Xuan He, Yanan Qiao

Remote Sensing · 06.08.2026 · 18 (15) · 2631

DOI: 10.3390/rs18152631 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 06.08.2026. Verfasst von Junge Bo und 3 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Hyperspectral image super-resolution (HSI SR) aims to recover high-resolution hyperspectral images from low-resolution observations while preserving spatial details and spectral fidelity. Accurate spectral preservation is a key distinction between HSI SR and natural image SR. Recent hybrid methods combining convolutional neural networks (CNNs) and Transformers have substantially improved spatial reconstruction performance, but spectral fidelity remains insufficiently explored. Our frequency perturbation analysis reveals that CNNs and Transformers exhibit complementary frequency-response characteristics. To explicitly exploit this complementarity, we propose a Frequency Alignment Dual-Branch Network (FAD-Net) for HSI SR. Specifically, the Multi-Scale Frequency Refinement Branch (MFRB) restores local high-frequency details through wavelet decomposition, while the Spatial–Spectral Frequency Interaction Branch (S2FIB) models long-range spatial and spectral dependencies via window-based attention. The two branches are fused by a Frequency Alignment Block (FAB). During training, Hyperspectral Frequency Loss and Spectral Angle Mapper loss are further introduced to constrain the reconstructed results toward the ground truth. Experiments on Chikusei, Botswana, and Pavia Center at ×2, ×3, and ×4 scales show that FAD-Net achieves the best Spectral Angle Mapper (SAM) in all nine settings while maintaining competitive spatial performance.

Zitieren

APA

Bo, J., Ling, C., He, J., & Qiao, Y. (2026). FAD-Net: Frequency Alignment Dual-Branch Network for Hyperspectral Image Super-Resolution. Remote Sensing, 18(15), 2631. https://doi.org/10.3390/rs18152631

Vancouver

Bo J, Ling C, He J, Qiao Y. FAD-Net: Frequency Alignment Dual-Branch Network for Hyperspectral Image Super-Resolution. Remote Sensing. 2026;18(15):2631. doi:10.3390/rs18152631

BibTeX
@article{Bo2026,
  author = {Bo, Junge and Ling, Chen and He, Ji-Xuan and Qiao, Yanan},
  title = {FAD-Net: Frequency Alignment Dual-Branch Network for Hyperspectral Image Super-Resolution},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {15},
  pages = {2631},
  doi = {10.3390/rs18152631}
}
Kostenlos starten →