Free-space and solid-matrix-media diffraction neural network masks made by two-photon lithography DOI

Tigran Baluian,

Daryana Pechkurova,

A. Konovalova

et al.

Published: Nov. 8, 2024

Neural networks are powerful tools for solving many modern problems. One of the options optical implementation a neural network is diffraction network, which consists one or several layers different-sized pixels on radiation diffracts. The pixel parameters tightly bound with desired wavelength. In this work, we printed masks range using two-photon laser lithography. Applying coordinate stabilization approach and preserving temperature humidity allowed to print up 10 nm height difference 2.3 average surface roughness.

Language: Английский

Advancements and Applications of Diffractive Optical Elements in Contemporary Optics: A Comprehensive Overview DOI
Svetlana N. Khonina, Nikolay L. Kazanskiy, Р. В. Скиданов

et al.

Advanced Materials Technologies, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 31, 2024

Abstract Diffractive optical elements (DOEs) represent a revolutionary advancement in modern optics, offering unparalleled versatility and efficiency various applications. Their significance lies their ability to manipulate light waves with intricate patterns, enabling functionalities beyond what traditional refractive optics can achieve. DOEs find widespread use fields such as laser beam shaping, holography, communications, imaging systems. By precisely controlling the phase amplitude of light, generate complex structures, correct aberrations, enhance performance Moreover, compact size, lightweight nature, potential for mass production make them indispensable designing efficient devices diverse industrial scientific From improving systems innovative display technologies, continue drive advancements promising even more exciting possibilities future. In this review, critical importance is illuminated explore profound implications contemporary era.

Language: Английский

Citations

12
Index-Matching Two-Photon Polymerization for Enhancing Machining Accuracy of Diffractive Neural Networks DOI Creative Commons

Michelle Fu,

Xiao‐Guang Ma, Weihong Shen

et al.

Photonics, Journal Year: 2025, Volume and Issue: 12(5), P. 473 - 473

Published: May 12, 2025

Two-photon polymerization (TPP) is an effective and rapid method for prototyping diffractive neural networks (DNNs). However, DNNs’ accuracy can be diminished by phase aberrations resulting from substrate misalignment in fabrication. To address this, we introduce index-matched two-photon (IM-TPP) fabricating DNNs. Numerical simulations show that on tilted substrates improved 91.50% to 95.00%. Experimentally, the IM-TPP process enhances device 3.00% (91.67% 94.67%), closely matching theoretical simulated of 95.03%. Additionally, average multiple batches samples reached 94.86%. reduces influence tilt error, improves performance manufacturing repeatability, provides a new high-precision optical computing elements.

Language: Английский

Citations

0
Synergy between AI and Optical Metasurfaces: A Critical Overview of Recent Advances DOI Creative Commons
Zoran Jakšić

Photonics, Journal Year: 2024, Volume and Issue: 11(5), P. 442 - 442

Published: May 9, 2024

The interplay between two paradigms, artificial intelligence (AI) and optical metasurfaces, nowadays appears obvious unavoidable. AI is permeating literally all facets of human activity, from science arts to everyday life. On the other hand, metasurfaces offer diverse sophisticated multifunctionalities, many which appeared impossible only a short time ago. use for optimization general approach that has become ubiquitous. However, here we are witnessing two-way process—AI improving but some also AI. helps design, analyze utilize while ensure creation all-optical chips. This ensures positive feedback where each enhances one: this may well be revolution in making. A vast number publications already cover either first or second direction; modest includes both. an attempt make reader-friendly critical overview emerging synergy. It succinctly reviews research trends, stressing most recent findings. Then, it considers possible future developments challenges. author hopes broad interdisciplinary will useful both dedicated experts scholarly audience.

Language: Английский

Citations

3
All-Optical Diffractive Deep Neural Networks Enabled Laser-Reduced Graphene Oxide Tactile Sensor for Braille Recognition DOI
Xing Liu, Fang Li, Fangyi Zhang

et al.

ACS Applied Electronic Materials, Journal Year: 2024, Volume and Issue: 6(3), P. 2049 - 2058

Published: March 15, 2024

All-optical diffractive deep neural networks (D2NNs) show a wide range of applications in image recognition and artificial vision due to their advantages high-speed parallel processing, low energy consumption, excellent anti-interference ability. However, there is relatively limited research applying D2NNs for tactile perception. In this study, we propose an automatic Braille method based on sensors. A flexible molybdenum disulfide-doped laser-reduced graphene oxide (LRGO/MoS2) sensor was fabricated with the laser direct writing method. The LRGO/MoS2 shows sensitivity 9.8 kPa–1, response/recovery time 0.14/0.10 s cyclic stability. can be employed capture character information real convert it into digital signals as inputs all-optical D2NNs. characters achieved five diffraction layers, system finally realize accuracy 100% recognition. strategy integrating sensors learning paves path realizing low-cost, fast, accurate, efficient system.

Language: Английский

Citations

1
Terahertz optical pattern recognition with rotation and scaling enhanced by a 3D-printed diffractive deep neural network DOI Creative Commons

Chenjie Xiong,

Xudong Wu,

Jianzhou Huang

et al.

Optics Express, Journal Year: 2024, Volume and Issue: 32(16), P. 27635 - 27635

Published: July 8, 2024

Optical pattern recognition (OPR) has the potential to be a valuable tool in field of terahertz (THz) imaging, with advantage being capable image single-point detection, which reduces overall system costs. However, this application is limited traditional OPR that rotation and scaling input will bring about an offset spot. Here we demonstrate full-diffractive method maintain spot at fixed position, even when rotated or scaled, by using all-optical diffractive deep neural network. The network composed two layers optical elements (DOEs) without 4f-system, 3D-printed all-in-one. Experimental results show our device can achieve stable regardless its (from 0° 360°) (with ratio from 1 1/1.9). This work expected provide enhanced functionality for compact THz systems imaging security applications.

Language: Английский

Citations

1
2bit Nonlinear Diffractive Deep Neural Network (2bit ND2NN): A quantized nonlinear all-optical diffractive deep neural network implementation DOI
Yichen Sun, Mingli Dong, Mingxin Yu

et al.

Optics & Laser Technology, Journal Year: 2024, Volume and Issue: 177, P. 111120 - 111120

Published: May 6, 2024

Language: Английский

Citations

0
A Variational Approach to Learning Photonic Unitary Operators DOI Creative Commons
Hadrian Bezuidenhout, Mwezi Koni, Jonathan Leach

et al.

Optics Express, Journal Year: 2024, Volume and Issue: 32(20), P. 35567 - 35567

Published: Sept. 4, 2024

Structured light, light tailored in its internal degrees of freedom, has become topical numerous quantum and classical information processing protocols. In this work, we harness the high dimensional nature structured modulated transverse spatial degree freedom to realize an adaptable scheme for learning unitary operations. Our approach borrows from concepts variational computing, where a search or optimization problem is mapped onto task finding minimum ground state energy given energy/goal function. We achieve by pseudo-random walk procedure over parameter space operation, implemented with optical matrix-vector multiplication enacted on arrays Gaussian modes exploiting partial Fourier transforming capabilities cylindrical lens measurement. outline concept theoretically, experimentally demonstrate that are able learn matrices dimensions d = 2, 4, 8, 16 average fidelities >90%. work advances can be adapted both process tomography unknown states channels.

Language: Английский

Citations

0
Free-space and solid-matrix-media diffraction neural network masks made by two-photon lithography DOI

Tigran Baluian,

Daryana Pechkurova,

A. Konovalova

et al.

Published: Nov. 8, 2024

Neural networks are powerful tools for solving many modern problems. One of the options optical implementation a neural network is diffraction network, which consists one or several layers different-sized pixels on radiation diffracts. The pixel parameters tightly bound with desired wavelength. In this work, we printed masks range using two-photon laser lithography. Applying coordinate stabilization approach and preserving temperature humidity allowed to print up 10 nm height difference 2.3 average surface roughness.

Language: Английский

Citations

0