Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group
Conferences > FiO > 2019 > JW4A > JW4A.101

Highly-parallel Optical Fourier Intensity Convolution Filter For Image Classification

Zibo Hu, Mario Miscuglio, Jonathan George, Yousra Alkabani, Tarek El Gazhawi, and Volker J. Sorger

Not Accessible

Your library or personal account may give you access

Abstract

Convolution Neural Networks (CNN) extracts features from large dataset by spatial filtering. We experimentally demonstrate convolutional filtering using Fourier optics. Unlike phase-based modulation, we show amplitude-based convolutions performing one layer of an artificial CNN.

© 2019 The Author(s)

PDF Article
More Like This
Free Space Optical Coprocessor for Image processing and Convolution Neural Network

Mario Miscuglio, Zibo Hu, Jonathan George, Zhizhen Ma, and Volker J. Sorger
SpT2E.6 Signal Processing in Photonic Communications (SPPCom) 2019

Million-channel parallelism Fourier-optic convolutional filter and neural network processor

Mario Miscuglio, Zibo Hu, Shurui Li, Jiaqi Gu, Aydin Babakhani, Puneet Gupta, Chee-Wei Wong, David Pan, Seth Bank, Hamed Dalir, and Volker J. Sorger
JF3A.4 CLEO: Applications and Technology (CLEO:A&T) 2020

Massive parallelism Fourier-optic convolutional processor

Mario Miscuglio, Zibo Hu, Shurui Li, Jiaqi Gu, Aydin Babakhani, Puneet Gupta, Chee-Wei Wong, David Pan, Seth Bank, Hamed Dalir, and Volker J. Sorger
SpTh3I.7 Signal Processing in Photonic Communications (SPPCom) 2020

View More...
Select as filters
Select Topics Cancel
Top
       
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.
Privacy | Terms of Use