Engineering and Technology

Ferroelectric photosensor network: an advanced hardware solution to real-time machine vision

B. Cui, Z. Fan, et al.

Discover an innovative self-powered computing paradigm revolutionizing real-time machine vision through a ferroelectric photosensor network (FE-PS-NET). This groundbreaking research, conducted by Boyuan Cui and colleagues, showcases the ability of ferroelectric photosensors to simultaneously capture and process images with remarkable precision.

00:00

Playback language: English

Index

Abstract

This paper introduces a self-powered in-sensor computing paradigm using a ferroelectric photosensor network (FE-PS-NET) for real-time machine vision applications. The FE-PS-NET, composed of ferroelectric photosensors (FE-PSs) with tunable photoresponsivities, allows simultaneous image capture and processing. Each FE-PS utilizes photovoltaic responses modulated by the remanent polarization of a Pb(Zr0.2Ti0.8)O3 layer, enabling the representation of signed weights in a single device. The interconnected FE-PSs function as an artificial neural network, performing in situ multiply-accumulate (MAC) operations. Experiments demonstrate successful real-time image processing, including binary classification and edge detection with 100% accuracy and F-Measure of 1, respectively.

Publisher

Nature Communications

Published On

Mar 31, 2022

Authors

Boyuan Cui, Zhen Fan, Wenjie Li, Yihong Chen, Shuai Dong, Zhengwei Tan, Shengliang Cheng, BoBo Tian, Ruiqiang Tao, Guo Tian, Deyang Chen, Zhipeng Hou, Minghui Qin, Min Zeng, Xubing Lu, Guofu Zhou, Xingsen Gao, Jun-Ming Liu

Related Publications

Explore these studies to deepen your understanding of the subject.

Computer Science

Real-time botnet detection on large network bandwidths using machine learning

J. Velasco-mata, V. González-castro, et al.

Medicine and Health

An integrated network representation of multiple cancer-specific data for graph-based machine learning

L. Pu, M. Singha, et al.

Veterinary Science

Divide-and-conquer: machine-learning integrates mammalian and viral traits with network features to predict virus-mammal associations

M. Wardeh, M. S. C. Blagrove, et al.

Physics

Nanosecond anomaly detection with decision trees and real-time application to exotic Higgs decays

S. T. Roche, Q. Bayer, et al.

Listen, Learn & Level Up

Over 10,000 hours of research content in 25+ fields, available in 12+ languages.

No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.

listen to research audio papers with researchbunny