Abstract

Computed tomography imaging spectrometry (CTIS) is a snapshot hyperspectral imaging technique that can obtain a three-dimensional (2D spatial $+$ 1D spectral) data cube of the scene captured within a single exposure. The CTIS inversion problem is typically highly ill-posed and is usually solved by time-consuming iterative algorithms. This work aims to take the full advantage of the recent advances in deep-learning algorithms to dramatically reduce the computational cost. For this purpose, a generative adversarial network is developed and integrated with self-attention, which cleverly exploits the clearly utilizable features of zero-order diffraction of CTIS. The proposed network is able to reconstruct a CTIS data cube (containing 31 spectral bands) in milliseconds with a higher quality than traditional methods and the state-of-the-art (SOTA). Simulation studies based on real image data sets confirmed the robustness and efficiency of the method. In numerical experiments with 1000 samples, the average reconstruction time for a single data cube was ${\sim}16\;{\rm{ms}}$. The robustness of the method against noise is also confirmed by numerical experiments with different levels of Gaussian noise. The CTIS generative adversarial network framework can be easily extended to solve CTIS problems with larger spatial and spectral dimensions, or migrated to other compressed spectral imaging modalities.

© 2023 Optica Publishing Group

Super-resolution computed tomography imaging spectrometry

Lei Yuan, Qiang Song, Hecong Liu, Kevin Heggarty, and Weiwei Cai
Photon. Res. 11(2) 212-224 (2023)

Dual-camera compressive hyperspectral imaging based on deep image prior and a guided filter

Haijun Tian, Jufeng Zhao, Junjie Zhu, Xuanji Tang, Guangmang Cui, and Changlun Hou
Appl. Opt. 62(14) 3649-3659 (2023)

On the value of CTIS imagery for neural-network-based classification: a simulation perspective

Clément Douarre, Carlos F. Crispim-Junior, Anthony Gelibert, Laure Tougne, and David Rousseau
Appl. Opt. 59(28) 8697-8710 (2020)

DH-GAN: a physics-driven untrained generative adversarial network for holographic imaging

Xiwen Chen, Hao Wang, Abolfazl Razi, Michael Kozicki, and Christopher Mann
Opt. Express 31(6) 10114-10135 (2023)

Hyperspectral image reconstruction via patch attention driven network

Yechuan Qiu, Shengjie Zhao, Xu Ma, Tong Zhang, and Gonzalo R. Arce
Opt. Express 31(12) 20221-20236 (2023)