Metasurface-Based Wavelength-Multiplexed Diffractive Neural Networks for Multilabel Intelligent Vision
nanoblocks with distinct geometries within each optical neuron, we achieve independent and parallel phase modulation across the RGB wavelengths for intelligent recognition of trichromatic objects. The proposed architecture is validated on a joint category-color Fashion-MNIST data set, achieving numerical and experimental accuracies of 91.7% and 85.8%, respectively. This work paves the way for high-dimensional intelligent perception systems through wavelength multiplexing.
Authors
- Daqi Zhu (ORCID: https://orcid.org/0000-0001-7252-4952)
- Rui Yang (ORCID: https://orcid.org/0000-0002-1886-5286)
- Yinan Zhang (ORCID: https://orcid.org/0000-0002-4538-4170)
- Mingke Wu
- Liangpeng Wei
- Lei Chen
- Zhao Wang
Institutions
- University of Shanghai for Science and Technology (CN)
Publication Details
- Journal
- Nano Letters
- Published
- 2026-04-21
- DOI
- https://doi.org/10.1021/acs.nanolett.6c00655
- Primary Topic
- Neural Networks and Reservoir Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China
- Science and Technology Commission of Shanghai Municipality
- Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning
- Frontiers Science Center for Shanghai Municipality