A Quantized Convolutional Neural Network Implemented With Memristor for Image Denoising and Recognition

Yuejun Zhang; Zhixin Wu; Zhixin Wu; Shuzhi Liu; Zhecheng Guo; Qilai Chen; Qilai Chen; Pingqi Gao; Pengjun Wang; Gang Liu

doi:10.3389/fnins.2021.717222

Frontiers in Neuroscience (Sep 2021)

A Quantized Convolutional Neural Network Implemented With Memristor for Image Denoising and Recognition

Yuejun Zhang,
Zhixin Wu,
Zhixin Wu,
Shuzhi Liu,
Zhecheng Guo,
Qilai Chen,
Qilai Chen,
Pingqi Gao,
Pengjun Wang,
Gang Liu

Affiliations

Yuejun Zhang: Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, China
Zhixin Wu: Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, China
Zhixin Wu: Department of Micro/Nano Electronics, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China
Shuzhi Liu: Department of Micro/Nano Electronics, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China
Zhecheng Guo: Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, China
Qilai Chen: Department of Micro/Nano Electronics, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China
Qilai Chen: School of Materials, Sun Yat-sen University, Guangzhou, China
Pingqi Gao: School of Materials, Sun Yat-sen University, Guangzhou, China
Pengjun Wang: College of Mathematics, Physics, and Electronic Information Engineering, Wenzhou University, Wenzhou, China
Gang Liu: Department of Micro/Nano Electronics, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China

DOI: https://doi.org/10.3389/fnins.2021.717222
Journal volume & issue: Vol. 15

Abstract

Read online

The interference of noise will cause the degradation of image quality, which can have a negative impact on the subsequent image processing and visual effect. Although the existing image denoising algorithms are relatively perfect, their computational efficiency is restricted by the performance of the computer, and the computational process consumes a lot of energy. In this paper, we propose a method for image denoising and recognition based on multi-conductance states of memristor devices. By regulating the evolution of Pt/ZnO/Pt memristor wires, 26 continuous conductance states were obtained. The image feature preservation and noise reduction are realized via the mapping between the conductance state and the image pixel. Furthermore, weight quantization of convolutional neural network is realized based on multi-conductance states. The simulation results show the feasibility of CNN for image denoising and recognition based on multi-conductance states. This method has a certain guiding significance for the construction of high-performance image noise reduction hardware system.

Published in Frontiers in Neuroscience

ISSN: 1662-4548 (Print); 1662-453X (Online)
Publisher: Frontiers Media S.A.
Country of publisher: Switzerland
LCC subjects: Medicine: Internal medicine: Neurosciences. Biological psychiatry. Neuropsychiatry
Website: http://www.frontiersin.org/neuroscience

About the journal

Abstract

Keywords