EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Mask-Guided Spatial–Spectral MLP Network for High-Resolution Hyperspectral Image Reconstruction

Xian-Hua Han iD, Jian Wang, Yen-Wei Chen iD

Sensors · 18.11.2024 · 24 (22) · 7362

DOI: 10.3390/s24227362 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 18.11.2024. Verfasst von Xian-Hua Han und 2 weiteren Autor:innen. Beteiligte Einrichtung: Graduate School of Artificial Intelligence and Science, Rikkyo University, Tokyo 171-8501, Japan. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Okawa Foundation for Information and Telecommunications Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Hyperspectral image (HSI) reconstruction is a critical and indispensable step in spectral compressive imaging (CASSI) systems and directly affects our ability to capture high-quality images in dynamic environments. Recent research has increasingly focused on deep unfolding frameworks for HSI reconstruction, showing notable progress. However, these approaches have to break the optimization task into two sub-problems, solving them iteratively over multiple stages, which leads to large models and high computational overheads. This study presents a simple yet effective method that passes the degradation information (sensing mask) through a deep learning network to disentangle the degradation and the latent target’s representations. Specifically, we design a lightweight MLP block to capture non-local similarities and long-range dependencies across both spatial and spectral domains, and investigate an attention-based mask modelling module to achieve the spatial–spectral-adaptive degradation representationthat is fed to the MLP-based network. To enhance the information flow between MLP blocks, we introduce a multi-level fusion module and apply reconstruction heads to different MLP features for deeper supervision. Additionally, we combine the projection loss from compressive measurements with reconstruction loss to create a dual-domain loss, ensuring consistent optical detection during HS reconstruction. Experiments on benchmark HS datasets show that our method outperforms state-of-the-art approaches in terms of both reconstruction accuracy and efficiency, reducing computational and memory costs.

Schlagwörter

Stichwörter der Autor:innen

MLP networkdegradationhyperspectral image reconstructionlong dependencysensing maskspatial–spectral modelling

Autor:innen und Institutionen

  • Graduate School of Artificial Intelligence and Science, Rikkyo University, Tokyo 171-8501, Japan.
  • School of Information Science and Engineering, Shandong Normal University, Jinan 250358, China.
  • College of Information Science and Engineering, Ritsumeikan University, Osaka 603-8577, Japan.

ORCID-Profile: Xian-Hua Han, Yen-Wei Chen

Förderung

  • Okawa Foundation for Information and Telecommunications Foundation

Zitieren

APA

Han, X. H., Wang, J., & Chen, Y. W. (2024). Mask-Guided Spatial–Spectral MLP Network for High-Resolution Hyperspectral Image Reconstruction. Sensors, 24(22), 7362. https://doi.org/10.3390/s24227362

Vancouver

Han XH, Wang J, Chen YW. Mask-Guided Spatial–Spectral MLP Network for High-Resolution Hyperspectral Image Reconstruction. Sensors. 2024;24(22):7362. doi:10.3390/s24227362

BibTeX
@article{Han2024,
  author = {Han, Xian-Hua and Wang, Jian and Chen, Yen-Wei},
  title = {Mask-Guided Spatial–Spectral MLP Network for High-Resolution Hyperspectral Image Reconstruction},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {22},
  pages = {7362},
  doi = {10.3390/s24227362}
}
Kostenlos starten →