期刊
ADVANCED THEORY AND SIMULATIONS
卷 4, 期 3, 页码 -出版社
WILEY-V C H VERLAG GMBH
DOI: 10.1002/adts.202000299
关键词
computational spectroscopy; deep learning; hyperspectral imaging; optical inverse design
资金
- National Key R&D Program of China [2018YFA0701400]
- Major Research Plan of the National Natural Science Foundation of China [92050115]
- Fundamental Research Funds for the Central Universities [2019QNA5006]
- ZJU-Sunny Photonics Innovation Center [2019-01]
- Zhejiang Lab [2020MC0AE01]
A neural network-based framework called parameter constrained spectral encoder and decoder (PCSED) is presented for designing BEST filters in spectroscopic instruments, linking mathematical optimum and practical limits. This approach results in a spectral camera based on BEST filters with higher reconstruction accuracy and better tolerance to fabrication errors. PCSED's generalizability is validated in designing metasurface- and interference-thin-film-based BEST filters.
Computational spectroscopic instruments with broadband encoding stochastic (BEST) filters allow the reconstruction of the spectrum at high precision with only a few filters. However, conventional design manners of BEST filters are often heuristic and may fail to fully explore the encoding potential of BEST filters. The parameter constrained spectral encoder and decoder (PCSED)-a neural network-based framework-is presented for the design of BEST filters in spectroscopic instruments. By incorporating the target spectral response definition and the optical design procedures comprehensively, PCSED links the mathematical optimum and practical limits confined by available fabrication techniques. Benefiting from this, a BEST-filter-based spectral camera presents a higher reconstruction accuracy with up to 30 times enhancement and a better tolerance to fabrication errors. The generalizability of PCSED is validated in designing metasurface- and interference-thin-film-based BEST filters.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据