Compressive dual-comb spectroscopy

Kawai, Maki; Kageyama, Takahiro; Horisaki, Ryoichi; Ideguchi, Takuro

doi:10.1038/s41598-021-93005-1

Cited by 9 publications

(2 citation statements)

References 33 publications

(53 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To mitigate this, researchers start to explore snapshot spectral imaging [10]. Early endeavors include integral field spectrometry, multispectral beam splitting, and image-replicating imaging spectrometer [11]. These methods, though achieve multispectral imaging through splitting light [12] [13], still fail to obtain massive spectral channels and require bulky optical systems.…”

Section: Introductionmentioning

confidence: 99%

RDFNet: Regional Dynamic FISTA-Net for Spectral Snapshot Compressive Imaging

Zhou,

Xu,

Dong

et al. 2023

Preprint

View full text Add to dashboard Cite

Deep convolutional neural networks have recently shown promising results in compressive spectral reconstruction. Previous methods, however, usually adopt a single mapping function for sparse representation. Considering that different regions have distinct characteristics, it is desirable to apply various mapping functions to adjust different regions' transformations dynamically. With this in mind, we first introduce a regional dynamic way of using Fast Iterative Shrinkage-Thresholding Algorithm (FISTA) to exploit regional characteristics and derive dynamic sparse representations. Then, we propose to unfold the process into a hierarchical dynamic deep network, dubbed RDFNet. The network comprises multiple regional dynamic blocks and corresponding pixel-wise adaptive soft-thresholding modules, respectively in charge of region-based dynamic mapping and pixel-wise soft-thresholding selection. The regional dynamic block guides the network to adjust the transformation domain for different regions. Equipped with the adaptive soft-thresholding, our proposed regional dynamic architecture can also learn appropriate shrinkage scale in a pixel-wise manner. Extensive experiments on both simulated and real data demonstrate that our method outperforms prior state-of-the-arts. Our code and data are available at https://github.com/SherryZhou97/RDFNet.

show abstract

Section: Introductionmentioning

confidence: 99%

RDFNet: Regional Dynamic FISTA-Net for Spectral Snapshot Compressive Imaging

Zhou,

Xu,

Dong

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…We explore two DCS modalities distinctively enabled by this approach: compressive and recurrence sampling. Compressive sensing has only been theoretically analyzed for DCS to show it could combine high speed and high frequency resolution 18,19 . The unique capability to program arbitrary random sampling pattern lets us experimentally demonstrate compressive sensing at compression factors up to 155 with a corresponding reduction in acquisition time while retrieving the absorption spectrum at the full spectral resolution of fr.…”

mentioning

confidence: 99%

Free-form dual-comb spectroscopy for compressive sensing and imaging

Giorgetta

Potvin²,

Deschênes³

et al. 2023

Preprint

View full text Add to dashboard Cite

Time programmable frequency combs enable new measurement paradigms for dual-comb spectroscopy (DCS) that are free of many of the constraints found in traditional DCS. As opposed to fixing the repetition rate offset between combs, free-form DCS is based on full control of the temporal offset between the dual comb pulse trains, thereby enabling user-selectable sampling patterns that optimize resolution, signal-to-noise ratio, species selectivity or acquisition time. Here we show that free-form DCS enables compressive sensing. We demonstrate compressive sensing at factors up to 155, with a corresponding reduction in acquisition time, while maintaining spectral resolution. We also demonstrate molecular recurrence sampling (which can be thought of as an extreme case of compressive sensing) for detection of a single gas with >15× higher sensitivity than traditional DCS. Finally, free-form DCS can more easily enable spectral imaging since its rf signal is narrow band, in contrast to traditional DCS, and therefore compatible with the finite readout times of focal plane arrays. We demonstrate imaging of methane plumes across a 128×64 pixel focal plane at a 250 Hz rate. In the future, this flexible free-form approach can enable applications ranging from rapid open-path spectroscopy to nonlinear multidimensional comb-based spectroscopy.

show abstract