EUVIMEDEUVIMED entdecken
Crossref

Parallel Spatial–Frequency Diffractive Neural Networks for Autonomous-Driving Visual Understanding

Peng Sun, Shu Li, Yuyang Duan, Yuchao Zhang, Tingxuan Li, Kun Tu, Yibo Dong, Min Gu, Jing Wang

ACS Photonics · 04.11.2026 · 13 (21) · 6422-6431

DOI: 10.1021/acsphotonics.6c01531 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACS Photonics, online erschienen am 26.10.2026. Verfasst von Peng Sun und 8 weiteren Autor:innen. Die Arbeit nennt 54 Literaturangaben. Gefördert von: Science and Technology Commission of Shanghai Municipality, Natural Science Foundation of Shanghai, Shanghai Frontiers Science Center Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Diffractive neural networks (DNNs) offer a promising route toward high-throughput, low-power optical computing; however, their application to autonomous-driving perception remains limited by insufficient representational capacity for complex scenes, owing to their relatively simple architectures and reliance on spatial-domain information alone. Here, we propose a spatial–frequency diffractive neural network (SFDNN) architecture that jointly exploits spatial- and frequency-domain information through two parallel sub-DNN branches. By leveraging the complementarity of these two domains, the proposed architecture improves feature separability and enhances perception performance, thereby outperforming conventional DNNs in the autonomous-driving segmentation task. Numerical results show that, under the same modulation-layer constraint, the SFDNN consistently surpasses single-domain models in segmentation tasks, improving reconstruction fidelity and spatial overlap over the best single-domain baseline by 10.9% and 23.2%, respectively. The framework further demonstrates strong generalization to traffic sign classification, achieving accuracies of up to 93.67%. Experimentally, we validate the proposed framework using a hybrid optical–digital system and demonstrate reliable target-region segmentation in a physical optical setup. Furthermore, a micrometer-scale SFDNN chip fabricated by laser nanoprinting confirms the feasibility of on-chip integration. This work provides a practical pathway to efficient, intelligent optical perception for autonomous driving and related edge-vision applications.

Förderung

  • Science and Technology Commission of Shanghai Municipality
  • Natural Science Foundation of Shanghai
  • Shanghai Frontiers Science Center Program
  • National Natural Science Foundation of China (NSFC)

Zitieren

APA

Sun, P., Li, S., Duan, Y., Zhang, Y., Li, T., Tu, K., Dong, Y., Gu, M., & Wang, J. (2026). Parallel Spatial–Frequency Diffractive Neural Networks for Autonomous-Driving Visual Understanding. ACS Photonics, 13(21), 6422-6431. https://doi.org/10.1021/acsphotonics.6c01531

Vancouver

Sun P, Li S, Duan Y, Zhang Y, Li T, Tu K, et al. Parallel Spatial–Frequency Diffractive Neural Networks for Autonomous-Driving Visual Understanding. ACS Photonics. 2026;13(21):6422-6431. doi:10.1021/acsphotonics.6c01531

BibTeX
@article{Sun2026,
  author = {Sun, Peng and Li, Shu and Duan, Yuyang and Zhang, Yuchao and Li, Tingxuan and Tu, Kun and Dong, Yibo and Gu, Min and Wang, Jing},
  title = {Parallel Spatial–Frequency Diffractive Neural Networks for Autonomous-Driving Visual Understanding},
  journal = {ACS Photonics},
  year = {2026},
  volume = {13},
  number = {21},
  pages = {6422--6431},
  doi = {10.1021/acsphotonics.6c01531}
}
Kostenlos starten →