Abstract
Unsupervised spectral reconstruction (SR) aims to recover the hyperspectral image (HSI) from corresponding RGB images without annotations. Existing SR methods achieve it from a single RGB image, hindered by the significant spectral distortion. Although several deep learning-based methods increase the SR accuracy by adding RGB images, their networks are always designed for other image recovery tasks, leaving huge room for improvement. To overcome this problem, we propose a novel, to our knowledge, approach that reconstructs the HSI from a pair of RGB images captured under two illuminations, significantly improving reconstruction accuracy. Specifically, an SR iterative model based on two illuminations is constructed at first. By unfolding the proximal gradient algorithm solving this SR model, an interpretable unsupervised deep network is proposed. All the modules in the proposed network have precise physical meanings, which enable our network to have superior performance and good generalization capability. Experimental results on two public datasets and our real-world images show the proposed method significantly improves both visually and quantitatively as compared with state-of-the-art methods.
© 2024 Optica Publishing Group
Full Article | PDF ArticleMore Like This
Xuheng Cao, Yusheng Lian, Zilong Liu, Han Zhou, Xiangmei Hu, Beiqing Huang, and Wan Zhang
Opt. Lett. 47(14) 3431-3434 (2022)
Kunshen Feng, Junfeng Li, Ming Li, Shilong Gao, Weiqi Deng, Haitao Xu, Jing Zhao, Yubin Lan, Yongbing Long, and Haidong Deng
Opt. Express 32(3) 4400-4412 (2024)
Zitong Ye, Xiaoyan Li, Yile Sun, Yuran Huang, Xu Liu, Yubing Han, and Cuifang Kuang
Opt. Lett. 49(9) 2205-2208 (2024)