Imaging through a scattering medium via model-driven deep learning
Optics Letters · 15.10.2023 · 48 (20) · 5285
DOI: 10.1364/ol.498796 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optics Letters, online erschienen am 04.10.2023. Verfasst von Fu Liu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 28 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Imaging through a scattering medium is of great significance in many areas. Especially, speckle correlation imaging has been valued for its noninvasiveness. In this work, we report a deep learning solution that incorporates the physical model and an additional regularization for high-fidelity speckle correlation imaging. Without large-scale data to train, the physical model and regularization prior provide a correct direction for neural network to precisely reconstruct hidden objects from speckle under different scattering scenarios and noise levels. Experimental results demonstrate that the proposed method presents a significant advance in improving generalization and combating the invasion of noise.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Liu, F., Meng, X., Yin, Y., & Yang, X. (2023). Imaging through a scattering medium via model-driven deep learning. Optics Letters, 48(20), 5285. https://doi.org/10.1364/ol.498796
Vancouver
Liu F, Meng X, Yin Y, Yang X. Imaging through a scattering medium via model-driven deep learning. Optics Letters. 2023;48(20):5285. doi:10.1364/ol.498796
BibTeX
@article{Liu2023,
author = {Liu, Fu and Meng, Xiangfeng and Yin, Yongkai and Yang, Xiulun},
title = {Imaging through a scattering medium via model-driven deep learning},
journal = {Optics Letters},
year = {2023},
volume = {48},
number = {20},
pages = {5285},
doi = {10.1364/ol.498796}
}
EUVIMED