Deep-learning based on-chip rapid spectral imaging with high spatial resolution  

作　　者：Jiawei Yang Kaiyu Cui Yidong Huang Wei Zhang Xue Feng Fang Liu 

机构地区：[1]Department of Electronic Engineering,Tsinghua University,Beijing 100084,China [2]Beijing National Research Center for Information Science and Technology(BNRist),Tsinghua University,Beijing 100084,China [3]Bejing Academy of Quantum Information Science,Beijing 100084,China

出　　处：《Chip》2023年第2期26-33,共8页芯片（英文）

基　　金：The National Natural Science Foundation of China(Grant No.U22A6004);The National Key Research and Development Program of China(2022YFF1501600).

摘　　要：Spectral imaging extends the concept of traditional color cameras to capture images across multiple spectral channels and has broad ap-plication prospects.Conventional spectral cameras based on scanning methods suffer from the drawbacks of low acquisition speed and large volume.On-chip computational spectral imaging based on metasur-face filters provides a promising scheme for portable applications,but endures long computation time due to point-by-point iterative spec-tral reconstruction and mosaic effect in the reconstructed spectral im-ages.In this study,on-chip rapid spectral imaging was demonstrated,which eliminated the mosaic effect in the spectral image by deep-learning-based spectral data cube reconstruction.The experimental results show that 4 orders of magnitude faster than the iterative spec-tral reconstruction were achieved,and the fidelity of the spectral re-construction for the standard color plate was over 99%for a standard color board.In particular,video-rate spectral imaging was demon-strated for moving objects and outdoor driving scenes with good per-formance for recognizing metamerism,where the concolorous sky and white cars can be distinguished via their spectra,showing great po-tential for autonomous driving and other practical applications in the field of intelligent perception.

关 键 词：Spectral imaging Deep learning Metasurface 

分 类 号：O433[机械工程—光学工程]