Skip to main navigation Skip to search Skip to main content

Freeform diffractive deep neural networks with intrinsically smooth layers plus a curvature constraint

  • Haijian Wang
  • , Zexin Feng*
  • , Yi Luo
  • *Corresponding author for this work
  • Beijing Institute of Technology
  • Tsinghua University

Research output: Contribution to journal › Article › peer-review

Abstract

Conventional diffractive deep neural networks (D2 NNs) rely on discretized phase distributions, which often lead to fabrication challenges, significant speckle noise, and sensitivity to perturbations. To overcome these limitations, we propose a D2 NN architecture based on smoothly freeform surfaces with C2 continuity. Specifically, we employ multi-level cubic B-splines to represent the phase distribution of each diffractive layer, which naturally yields a smooth phase profile while preserving high design freedom. An explicit curvature constraint is further imposed to regularize the surface curvature, which facilitates fabrication by, e.g., diamond turning. Results show that, compared with a discrete-phase D2 NN, our method even achieves a slightly higher accuracy of 95.97% on MNIST, while being more robust against alignment errors.

Original languageEnglish
Pages (from-to)4020-4023
Number of pages4
JournalOptics Letters
Volume51
Issue number14
DOIs
Publication statusPublished - 2026
Externally publishedYes

Fingerprint

Dive into the research topics of 'Freeform diffractive deep neural networks with intrinsically smooth layers plus a curvature constraint'. Together they form a unique fingerprint.

Cite this