Computational Hyperspectral Imaging with Diffractive Optics and Deep Residual Network

Hyperspectral imaging critically serves for various fields such as remote sensing, biomedical and agriculture. Its potential can be exploited to a greater extent when combined with deep learning methods, which improve the reconstructed hyperspectral image quality and reduce the processing time. In this paper, we propose a novel snapshot hyperspectral imaging system using optimized diffractive optical element and color filter along with the residual dense network. We evaluate our method through simulations considering the effects of each optical element and noise. Simulation results demonstrate high-quality hyperspectral image reconstruction capabilities through the proposed computational hyperspectral camera.