End-to-end metasurface inverse design for single-shot multi-channel imaging
<jats:p>We introduce end-to-end inverse design for multi-channel imaging, in which a nanophotonic frontend is optimized in conjunction with an image-processing backend to extract depth, spectral and polarization channels from a single monochrome image. Unlike diffractive optics, we show that s...
| Main Authors: | , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
Optica Publishing Group
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/1721.1/143936 |
| _version_ | 1811080528956227584 |
|---|---|
| author | Lin, Zin Pestourie, Raphaël Roques-Carmes, Charles Li, Zhaoyi Capasso, Federico Soljačić, Marin Johnson, Steven G. |
| author2 | Massachusetts Institute of Technology. Department of Mathematics |
| author_facet | Massachusetts Institute of Technology. Department of Mathematics Lin, Zin Pestourie, Raphaël Roques-Carmes, Charles Li, Zhaoyi Capasso, Federico Soljačić, Marin Johnson, Steven G. |
| author_sort | Lin, Zin |
| collection | MIT |
| description | <jats:p>We introduce end-to-end inverse design for multi-channel imaging, in which a nanophotonic frontend is optimized in conjunction with an image-processing backend to extract depth, spectral and polarization channels from a single monochrome image. Unlike diffractive optics, we show that subwavelength-scale “metasurface” designs can easily distinguish similar wavelength and polarization inputs. The proposed technique integrates a single-layer metasurface frontend with an efficient Tikhonov reconstruction backend, without any additional optics except a grayscale sensor. Our method yields multi-channel imaging by spontaneous demultiplexing: the metaoptics front-end separates different channels into distinct spatial domains whose locations on the sensor are optimally discovered by the inverse-design algorithm. We present large-area metasurface designs, compatible with standard lithography, for multi-spectral imaging, depth-spectral imaging, and “all-in-one” spectro-polarimetric-depth imaging with robust reconstruction performance (≲ 10% error with 1% detector noise). In contrast to neural networks, our framework is physically interpretable and does not require large training sets. It can be used to reconstruct arbitrary three-dimensional scenes with full multi-wavelength spectra and polarization textures.</jats:p> |
| first_indexed | 2024-09-23T11:33:09Z |
| format | Article |
| id | mit-1721.1/143936 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T11:33:09Z |
| publishDate | 2022 |
| publisher | Optica Publishing Group |
| record_format | dspace |
| spelling | mit-1721.1/1439362023-02-08T19:47:48Z End-to-end metasurface inverse design for single-shot multi-channel imaging Lin, Zin Pestourie, Raphaël Roques-Carmes, Charles Li, Zhaoyi Capasso, Federico Soljačić, Marin Johnson, Steven G. Massachusetts Institute of Technology. Department of Mathematics Massachusetts Institute of Technology. Research Laboratory of Electronics Massachusetts Institute of Technology. Department of Physics Atomic and Molecular Physics, and Optics <jats:p>We introduce end-to-end inverse design for multi-channel imaging, in which a nanophotonic frontend is optimized in conjunction with an image-processing backend to extract depth, spectral and polarization channels from a single monochrome image. Unlike diffractive optics, we show that subwavelength-scale “metasurface” designs can easily distinguish similar wavelength and polarization inputs. The proposed technique integrates a single-layer metasurface frontend with an efficient Tikhonov reconstruction backend, without any additional optics except a grayscale sensor. Our method yields multi-channel imaging by spontaneous demultiplexing: the metaoptics front-end separates different channels into distinct spatial domains whose locations on the sensor are optimally discovered by the inverse-design algorithm. We present large-area metasurface designs, compatible with standard lithography, for multi-spectral imaging, depth-spectral imaging, and “all-in-one” spectro-polarimetric-depth imaging with robust reconstruction performance (≲ 10% error with 1% detector noise). In contrast to neural networks, our framework is physically interpretable and does not require large training sets. It can be used to reconstruct arbitrary three-dimensional scenes with full multi-wavelength spectra and polarization textures.</jats:p> 2022-07-21T16:27:06Z 2022-07-21T16:27:06Z 2022-07-19 Article http://purl.org/eprint/type/JournalArticle 1094-4087 https://hdl.handle.net/1721.1/143936 Lin, Zin, Pestourie, Raphaël, Roques-Carmes, Charles, Li, Zhaoyi, Capasso, Federico et al. 2022. "End-to-end metasurface inverse design for single-shot multi-channel imaging." 30 (16). 10.1364/oe.449985 Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Optica Publishing Group Prof. Johnson |
| spellingShingle | Atomic and Molecular Physics, and Optics Lin, Zin Pestourie, Raphaël Roques-Carmes, Charles Li, Zhaoyi Capasso, Federico Soljačić, Marin Johnson, Steven G. End-to-end metasurface inverse design for single-shot multi-channel imaging |
| title | End-to-end metasurface inverse design for single-shot multi-channel imaging |
| title_full | End-to-end metasurface inverse design for single-shot multi-channel imaging |
| title_fullStr | End-to-end metasurface inverse design for single-shot multi-channel imaging |
| title_full_unstemmed | End-to-end metasurface inverse design for single-shot multi-channel imaging |
| title_short | End-to-end metasurface inverse design for single-shot multi-channel imaging |
| title_sort | end to end metasurface inverse design for single shot multi channel imaging |
| topic | Atomic and Molecular Physics, and Optics |
| url | https://hdl.handle.net/1721.1/143936 |
| work_keys_str_mv | AT linzin endtoendmetasurfaceinversedesignforsingleshotmultichannelimaging AT pestourieraphael endtoendmetasurfaceinversedesignforsingleshotmultichannelimaging AT roquescarmescharles endtoendmetasurfaceinversedesignforsingleshotmultichannelimaging AT lizhaoyi endtoendmetasurfaceinversedesignforsingleshotmultichannelimaging AT capassofederico endtoendmetasurfaceinversedesignforsingleshotmultichannelimaging AT soljacicmarin endtoendmetasurfaceinversedesignforsingleshotmultichannelimaging AT johnsonsteveng endtoendmetasurfaceinversedesignforsingleshotmultichannelimaging |