Metasurface-generated large and arbitrary analog convolution kernels for accelerated machine vision
Authors:
Preprint
Abstract
In the rapidly evolving field of artificial intelligence, convolutional neural networks are essential for tackling complex challenges such as machine vision and medical diagnosis. Recently, to address the challenges in processing speed and power consumption of conventional digital convolution operations, many optical components have been suggested to replace the digital convolution layer in the neural network, accelerating various machine vision tasks. Nonetheless, the analog nature of the optical convolution kernel has not been fully explored. Here, we develop a spatial frequency domain training method to create arbitrarily shaped analog convolution kernels using an optical metasurface as the convolution layer, with its receptive field largely surpassing digital convolution kernels. By employing spatial multiplexing, the multiple parallel convolution kernels with both positive and negative weights are generated under the incoherent illumination condition. We experimentally demonstrate a 98.59% classification accuracy on the MNIST dataset, with simulations showing 92.63% and 68.67% accuracy on the Fashion-MNIST and CIFAR-10 datasets with additional digital layers. This work underscores the unique advantage of analog optical convolution, offering a promising avenue to accelerate machine vision tasks, especially in edge devices.Keywords
All Related Versions
This publication has 45 references indexed in Scilit:
- Meta‐Optics Based Parallel Convolutional Processing for Neural Network AcceleratorLaser & Photonics Reviews, 2024
- Metalens for Accelerated Optoelectronic Edge Detection under Ambient IlluminationNano Letters, 2023
- Integrated Photonic Neural Networks: Opportunities and ChallengesACS Photonics, 2023
- Angular Transmission Response of In-Plane Symmetry-Breaking Quasi-BIC All-Dielectric MetasurfacesACS Photonics, 2022
- Physics‐Aware Machine Learning and Adversarial Attack in Complex‐Valued Reconfigurable Diffractive All‐Optical Neural NetworkLaser & Photonics Reviews, 2022
- Metasurface-enabled on-chip multiplexed diffractive neural networks in the visibleLight: Science & Applications, 2022
- LOEN: Lensless opto-electronic neural network empowered machine visionLight: Science & Applications, 2022
- Ultracompact meta-imagers for arbitrary all-optical convolutionLight: Science & Applications, 2022
- Single-Layer Metasurface with Controllable Multiwavelength FunctionsNano Letters, 2018
- Deep learning with coherent nanophotonic circuitsNature Photonics, 2017