Tailoring the optical transfer function of nonlocal metasurfaces for targeted image processing via an automated inverse design framework

Abstract

Nonlocal metasurfaces exhibit significant potential for advanced all-optical image processing by leveraging their exceptional capability to regulate spatial dispersion through precise tailoring of optical transfer functions (OTFs). However, the inverse design of specific OTFs remains challenging due to the inherently complex and highly nonlinear relationship between metasurface structural parameters and angular-dependent optical responses, which conventional empirical trial-and-error approaches struggle to address. To overcome this limitation, we propose an automated inverse design framework integrating a deep neural network acting as a forward predictor with Bayesian optimization. This framework enables automated OTF tailoring by optimizing metasurface structural parameters for targeted image processing operations at desired wavelengths within the 1,200-1,400 nm range. We validate the framework by designing nine dedicated silicon hollow brick metasurfaces: for each operational wavelength (1,250, 1,300, and 1,350 nm), three distinct devices are engineered to separately execute 2D second-order differentiation, 2D fourth-order differentiation, and 2D Gaussian high-pass filtering in transmission mode through targeted OTF engineering. These inversely designed nonlocal metasurfaces achieve a numerical aperture close to 0.4 and serve as fundamental components for edge detection and image sharpening. This intelligent, automated design paradigm dramatically accelerates the design process and significantly expands the scope of achievable functionalities for optical computing metasurfaces, paving the way for more sophisticated all-optical information processing systems.

Keywords

Optical transfer function / automated inverse design / optical image processing / nonlocal metasurface

References

[1] Mcmahon PL.The physics of optical computing.Nat Rev Phys2023;5:717-34

[2] Xu Z,Ma M,Dai Q.Large-scale photonic chiplet Taichi empowers 160-TOPS/W artificial general intelligence.Science2024;384:202-9

[3] Hua S,Yu S.An integrated large-scale photonic accelerator with ultralow latency.Nature2025;640:361-7 PMCID:PMC11981923

[4] Chamoli SK,Fan Y.Nonlocal flat optics for size-selective image processing and denoising.Nat Commun2025;16:4473 PMCID:PMC12078730

[5] Zhou Y,Kravchenko II.Flat optics for image differentiation.Nat Photonics2020;14:316-23

[6] Zangeneh-nejad F,Alù A.Analogue computing with metamaterials.Nat Rev Mater2020;6:207-25

[7] Silva A,Castaldi G,Alù A.Performing mathematical operations with metamaterials.Science2014;343:160-3

[8] Goodman JW. Introduction to fourier optics; Englewood: Roberts and Company publishers; 2005. Available from: https://books.google.com/books?hl=en&lr=&id=ow5xs_Rtt9AC&oi=fnd&pg=PR7&dq=Introduction+to+Fourier+optics&ots=G_m2BK4GMK&sig=bQx1xvmNt3X_O9oHZZiuL1SM5KI#v=onepage&q=Introduction%20to%20Fourier%20optics&f=false [Last accessed on 20 Mar 2026]

[9] Dorrah AH.Tunable structured light with flat optics.Science2022;376:eabi6860

[10] Qiu X,Fan Y,Chen L.Metasurface enabled high-order differentiator.Nat Commun2025;16:2437 PMCID:PMC11897169

[11] Tanriover I,Aydin K.Metasurface enabled broadband all optical edge detection in visible frequencies.Nat Commun2023;14:6484 PMCID:PMC10576829

[12] Yang G,Lee JS.Nonlocal phase-change metaoptics for reconfigurable nonvolatile image processing.Light Sci Appl2025;14:182 PMCID:PMC12053629

[13] Chen Y,Seppecher P,Wegener M.Nonlocal metamaterials and metasurfaces.Nat Rev Phys2025;7:299-312

[14] Zhou C,Li Y.Laplace differentiator based on metasurface with toroidal dipole resonance.Adv Funct Mater2024;34:2313777

[15] Shastri K.Nonlocal flat optics.Nat Photonics2022;17:36-47

[16] Kwon H,Cordaro A,Alù A.Nonlocal metasurfaces for optical signal processing.Phys Rev Lett2018;121:173004

[17] Kwon H,Sounas D,Alù A.Dual-polarization analog 2D image processing with nonlocal metasurfaces.ACS Photonics2020;7:1799-805

[18] Long OY,Jin W.Polarization-independent isotropic nonlocal metasurfaces with wavelength-controlled functionality.Phys Rev Appl2022;17:024029

[19] Cotrufo M,Singh S.Dispersion engineered metasurfaces for broadband, high-NA, high-efficiency, dual-polarization analog image processing.Nat Commun2023;14:7078 PMCID:PMC10625611

[20] Chen A.Dielectric nonlocal metasurfaces for fully solid-state ultrathin optical systems.ACS Photonics2021;8:1439-47

[21] Abdollahramezani S,Adibi A.Meta-optics for spatial optical analog computing.Nanophotonics2020;9:4075-95

[22] Qian C,Chen H.A guidance to intelligent metamaterials and metamaterials intelligence.Nat Commun2025;16:1154 PMCID:PMC11779837

[23] Chen MK,Sun Y.Artificial intelligence in meta-optics.Chem Rev2022;122:15356-413 PMCID:PMC9562283

[24] Li Z,Lin Z,Capasso F.Empowering metasurfaces with inverse design: principles and applications.ACS Photonics2022;9:2178-92

[25] Jiang J,Fan JA.Deep neural networks for the evaluation and design of photonic devices.Nat Rev Mater2020;6:679-700

[26] Li W,Tsvetkov D.Machine learning for engineering meta‐atoms with tailored multipolar resonances.Laser Photonics Rev2024;18:2300855

[27] Chi H,Ou X.Neural network-assisted end-to-end design for full light field control of meta-optics.Adv Mater2025;37:e2419621

[28] Li Y,Jiang M.Self-learning perfect optical chirality via a deep neural network.Phys Rev Lett2019;123:213902

[29] Tao C,Li Y,Zhou J.Efficient excitation of acoustic graphene plasmons for sub-nanoscale infrared sensing.J Opt Soc Am B2024;41:2280

[30] An S,Zheng B.A deep learning approach for objective-driven all-dielectric metasurface design.ACS Photonics2019;6:3196-207

[31] Makandar A.Image enhancement techniques using highpass and lowpass filters.Int J Comput Appl2015;109:21-7

[32] Xu L,Ma Y.Enhanced light-matter interactions in dielectric nanostructures via machine-learning approach.Adv Photon2020;2:1

[33] An S,Tang H.Multifunctional metasurface design with a generative adversarial network.Adv Opt Mater2021;9:2001433

[34] Ma W,Xiong B.Pushing the limits of functionality-multiplexing capability in metasurface design based on statistical machine learning.Adv Mater2022;34:e2110022

[35] Zhang N,Wang R.Deep-learning empowered customized chiral metasurface for calibration-free biosensing.Adv Mater2025;37:e2411490

[36] Cotrufo M,Arora A,Alù A.Polarization imaging and edge detection with image-processing metasurfaces.Optica2023;10:1331