Deep learning enabled real time speckle recognition and hyperspectral imaging using a multimode fiber array - Archive ouverte HAL
Article Dans Une Revue Optics Express Année : 2019

Deep learning enabled real time speckle recognition and hyperspectral imaging using a multimode fiber array

Résumé

We demonstrate the use of deep learning for fast spectral deconstruction of speckle patterns. The artificial neural network can be effectively trained using numerically constructed multispectral datasets taken from a measured spectral transmission matrix. Optimized neural networks trained on these datasets achieve reliable reconstruction of both discrete and continuous spectra from a monochromatic camera image. Deep learning is compared to analytical inversion methods as well as to a com-pressive sensing algorithm and shows favourable characteristics both in the oversampling and in the sparse undersampling (compressive) regimes. The deep learning approach offers significant advantages in robustness to drift or noise and in reconstruction speed. In a proof-of-principle demonstrator we achieve real time recovery of hyperspectral information using a multi-core, multi-mode fiber array as a random scattering medium.
Fichier principal
Vignette du fichier
1904.04673.pdf (7.81 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02484218 , version 1 (19-02-2020)

Identifiants

Citer

Ulas Kürüm, Peter Wiecha, Rebecca French, Otto L. Muskens. Deep learning enabled real time speckle recognition and hyperspectral imaging using a multimode fiber array. Optics Express, 2019, 27 (15), pp.20965. ⟨10.1364/OE.27.020965⟩. ⟨hal-02484218⟩
75 Consultations
72 Téléchargements

Altmetric

Partager

More