(PD) 2 : physics-driven phase diversity neural network for wavefront sensing
Optics Express · 27.07.2026 · 34 (15) · 28728
DOI: 10.1364/oe.606688 ↗Kurz gefasst
Fachartikel in der Zeitschrift Optics Express, online erschienen am 24.07.2026. Verfasst von Qixiang Gao und 3 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Gefördert von: Chang Guang Satellite Technology Co., Ltd..
Abstract (Originalsprache)
Phase-diversity wavefront sensing reconstructs aberrations from focused and defocused images without dedicated sensors, but it remains a high-dimensional nonlinear inverse problem. Conventional iterative methods are sensitive to initialization, noise, and model mismatch, whereas purely data-driven methods depend heavily on labeled data and often suffer from simulation-to-real gaps. This work focuses on point-source phase-diversity wavefront sensing and proposes a physics-driven self-supervised neural-network-based solution that estimates wavefront parameters from focused and defocused observations using image consistency under the forward physical model, without requiring large amounts of real labels. Simulations and real-world experiments show improved accuracy and robustness over L-BFGS and a purely data-driven baseline under the evaluated settings; under the ideal clean simulation setting, the proposed method reduces the wavefront residual RMS to 0.0058 λ and achieves a success rate of 98.40%.
Förderung
- Chang Guang Satellite Technology Co., Ltd.
Zitieren
APA
Gao, Q., Ma, Y., Li, Y., & Zhong, X. (2026). (PD) 2 : physics-driven phase diversity neural network for wavefront sensing. Optics Express, 34(15), 28728. https://doi.org/10.1364/oe.606688
Vancouver
Gao Q, Ma Y, Li Y, Zhong X. (PD) 2 : physics-driven phase diversity neural network for wavefront sensing. Optics Express. 2026;34(15):28728. doi:10.1364/oe.606688
BibTeX
@article{Gao2026,
author = {Gao, Qixiang and Ma, Yi and Li, Yunxiao and Zhong, Xing},
title = {(PD) 2 : physics-driven phase diversity neural network for wavefront sensing},
journal = {Optics Express},
year = {2026},
volume = {34},
number = {15},
pages = {28728},
doi = {10.1364/oe.606688}
}
EUVIMED