Noise-Robust Loop-Based Deep Optical Convolutional Neural Network
Photonics · 04.06.2026 · 13 (6) · 552
DOI: 10.3390/photonics13060552 ↗Kurz gefasst
Fachartikel in der Zeitschrift Photonics, online erschienen am 04.06.2026. Verfasst von Maryam Dehbashizadeh Chehreghan und 1 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: EC KDT JU CLEVER project.
Abstract (Originalsprache)
We demonstrate a loop-based deep optical convolutional neural network that reuses a single free-space optical hardware to realize network depth through repeated passes. Convolution is implemented with programmable SLM with Fourier plane kernels, nonlinearity is provided by the photorefractive phase-only response of a BSO crystal and converted to an effective intensity activation via spatial filtering, and pooling is performed optically using demagnified imaging with an iris. On MNIST, the BSO-based nonlinearity improves test accuracy from 90.8% (linear) to 95.7%, with optimal operation. We model realistic optical noises (laser fluctuation, aberration, detector misalignment, and dust) and compare them using an SSIM-normalized severity metric. Under noise at (s = 0.35) on Fashion-MNIST, accuracy drops from 88.53% (clean) to 79.5% (noisy inference); a feature-level noise-aware training strategy recovers performance to 86.87%. Together, these advances demonstrate that a compact, loop-based hybrid DOCNN, completed with simple optical nonlinearities, simplified pooling, and noise-aware learning, can improve accuracy under realistic conditions.
Förderung
- EC KDT JU CLEVER project
Zitieren
APA
Chehreghan, M. D., & Stabile, R. (2026). Noise-Robust Loop-Based Deep Optical Convolutional Neural Network. Photonics, 13(6), 552. https://doi.org/10.3390/photonics13060552
Vancouver
Chehreghan MD, Stabile R. Noise-Robust Loop-Based Deep Optical Convolutional Neural Network. Photonics. 2026;13(6):552. doi:10.3390/photonics13060552
BibTeX
@article{Chehreghan2026,
author = {Chehreghan, Maryam Dehbashizadeh and Stabile, Ripalta},
title = {Noise-Robust Loop-Based Deep Optical Convolutional Neural Network},
journal = {Photonics},
year = {2026},
volume = {13},
number = {6},
pages = {552},
doi = {10.3390/photonics13060552}
}
EUVIMED