A speckle-based approach to compressive hyperspectral imaging
A speckle-based approach to compressive hyperspectral imaging
Incorporating wavelength information into a monochrome image is of great interest for imaging spectroscopy. Here, we show that by exploiting the properties of multiple scattering materials, we can encode spectral information in a CMOS image, where the spectral resolution obtained is only limited by the scattering strength of the material. As a proof-of-concept, we demonstrate this technique using a thin multiple scattering layer of gallium phosphide nanowires and a microlens array. We achieve a spectral resolution of approximately 4 nm and a reconstructed image containing 64 pixels. We demonstrate that a computational technique which is commonly used in compressive sensing can be used to reconstruct both sparse and dense spectra, when undersampling and oversampling a signal, respectively. This method provides an ultra-compact solution to obtaining both spatial and spectral information in one measurement, for potential use in portable spectroscopy.
compressive sensing, hyperspectral imaging, multiple scattering, speckle
1-8
French, Rebecca
d6d6a85a-e351-4cc8-ae4a-827c35fe6b64
Gigan, Sylvain
f2f2026a-fbe0-4fbe-b3a1-d8f7b35cdd58
Muskens, Otto L.
2284101a-f9ef-4d79-8951-a6cda5bfc7f9
French, Rebecca
d6d6a85a-e351-4cc8-ae4a-827c35fe6b64
Gigan, Sylvain
f2f2026a-fbe0-4fbe-b3a1-d8f7b35cdd58
Muskens, Otto L.
2284101a-f9ef-4d79-8951-a6cda5bfc7f9
French, Rebecca, Gigan, Sylvain and Muskens, Otto L.
(2018)
A speckle-based approach to compressive hyperspectral imaging.
In Next-Generation Spectroscopic Technologies XI.
vol. 10657,
SPIE.
.
(doi:10.1117/12.2303993).
Record type:
Conference or Workshop Item
(Paper)
Abstract
Incorporating wavelength information into a monochrome image is of great interest for imaging spectroscopy. Here, we show that by exploiting the properties of multiple scattering materials, we can encode spectral information in a CMOS image, where the spectral resolution obtained is only limited by the scattering strength of the material. As a proof-of-concept, we demonstrate this technique using a thin multiple scattering layer of gallium phosphide nanowires and a microlens array. We achieve a spectral resolution of approximately 4 nm and a reconstructed image containing 64 pixels. We demonstrate that a computational technique which is commonly used in compressive sensing can be used to reconstruct both sparse and dense spectra, when undersampling and oversampling a signal, respectively. This method provides an ultra-compact solution to obtaining both spatial and spectral information in one measurement, for potential use in portable spectroscopy.
This record has no associated files available for download.
More information
e-pub ahead of print date: 14 May 2018
Venue - Dates:
Next-Generation Spectroscopic Technologies XI 2018, , Orlando, United States, 2018-04-16 - 2018-04-18
Keywords:
compressive sensing, hyperspectral imaging, multiple scattering, speckle
Identifiers
Local EPrints ID: 422926
URI: http://eprints.soton.ac.uk/id/eprint/422926
PURE UUID: 607207a5-2dbd-419b-be9a-e871008ae8c5
Catalogue record
Date deposited: 08 Aug 2018 16:30
Last modified: 16 Mar 2024 04:01
Export record
Altmetrics
Contributors
Author:
Rebecca French
Author:
Sylvain Gigan
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics