PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 12, 2026IEEE Transactions on Image Processing1 citations

Learning Retinex Prior for Compressive Hyperspectral Image Reconstruction

View Full Paper
MLMengzu LiuJXJunwei XuZYZhangxi Yan

Key Points

  • This research aims to enhance the quality of hyperspectral image reconstruction from compressed measurements by applying the Retinex prior.
  • Developed the Retinex Prior-Driven Unfolding Network (RPDUN) for image reconstruction.
  • Incorporated the Retinex prior as a regularization term in a multi-stage network.
  • Introduced the Adaptive Token Selection Transformer (ATST) to reduce noise and artifacts.
  • Conducted extensive experiments with simulated and real-world datasets.
  • RPDUN achieved state-of-the-art performance in hyperspectral image reconstruction.
  • Significantly improved reconstruction quality compared to existing methods.
  • Maintained computational efficiency during the process.

Abstract

Image reconstruction in coded aperture snapshot spectral compressive imaging (CASSI) aims to recover high-fidelity hyperspectral images (HSIs) from compressed 2D measurements. While deep unfolding networks have shown promising performance, the degradation induced by the CASSI degradation model often introduces global illumination discrepancies in the reconstructions, creating artifacts similar to those in low-light images. To address these challenges, we propose a novel Retinex Prior-Driven Unfolding Network (RPDUN), which unfolds the optimization incorporating the Retinex prior as a regularization term into a multi-stage network. This design provides global illumination adjustment for compressed measurements, effectively compensating for spatial-spectral degradation according to physical modulation and capturing intrinsic spectral characteristics. To the best of our knowledge, this is the first application of the Retinex prior in hyperspectral image reconstruction. Furthermore, to mitigate the noise in the reflectance domain, which can be amplified during decomposition, we introduce an Adaptive Token Selection Transformer (ATST). This module adaptively filters out weakly correlated tokens before the self-attention computation, effectively reducing noise and artifacts within the recovered reflectance map. Extensive experiments on both simulated and real-world datasets demonstrate that RPDUN achieves new state-of-the-art performance, significantly improving reconstruction quality while maintaining computational efficiency. The code is available at https://github.com/ZUGE0312/RPDUN.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Liu et al. (2026) studied this question.

synapsesocial.com/papers/698d6dd15be6419ac0d52f94https://doi.org/10.1109/tip.2026.3659746
Ask AI
Helpful
Bookmark
Share
View Full Paper