Snapshot spectral imaging based on aberration model-driven deep learning
Optics Letters · 01.06.2024 · 49 (11) · 2894
DOI: 10.1364/ol.523832 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optics Letters, online erschienen am 16.05.2024. Verfasst von Qiuyu Yue und 3 weiteren Autor:innen. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 16 Literaturangaben. Gefördert von: National Key Research and Development Program of China.
Abstract (Originalsprache)
Coded aperture snapshot spectral imaging (CASSI) can capture hyperspectral images (HSIs) in one shot, but it suffers from optical aberrations that degrade the reconstruction quality. Existing deep learning methods for CASSI reconstruction lose some performance on real data due to aberrations. We propose a method to restore high-resolution HSIs from a low-resolution CASSI measurement. We first generate realistic training data that mimics the optical aberrations of CASSI using a spectral imaging simulation technique. A generative network is then trained on this data to recover HSIs from a blurred and distorted CASSI measurement. Our method adapts to the optical system degradation model and thus improves the reconstruction robustness. Experiments on both simulated and real data indicate that our method significantly enhances the image quality of reconstruction outcomes and can be applied to different CASSI systems.
Förderung
- National Key Research and Development Program of China
Zitieren
APA
Yue, Q., Chen, B., Liu, X., & Zheng, Z. (2024). Snapshot spectral imaging based on aberration model-driven deep learning. Optics Letters, 49(11), 2894. https://doi.org/10.1364/ol.523832
Vancouver
Yue Q, Chen B, Liu X, Zheng Z. Snapshot spectral imaging based on aberration model-driven deep learning. Optics Letters. 2024;49(11):2894. doi:10.1364/ol.523832
BibTeX
@article{Yue2024,
author = {Yue, Qiuyu and Chen, Bingliang and Liu, Xinyu and Zheng, Zhenrong},
title = {Snapshot spectral imaging based on aberration model-driven deep learning},
journal = {Optics Letters},
year = {2024},
volume = {49},
number = {11},
pages = {2894},
doi = {10.1364/ol.523832}
}
EUVIMED