摘要
该文构建一种新的忆阻细胞神经网络,改进传统的忆阻突触桥电路,使之除了具有传统突触桥电路的优势外,还具有更加简化的电路和简化的权值变化条件.通过仿真电路模拟器(SPICE)仿真模拟该突触电路能够实现权值运算.将忆阻细胞神经网络用于图像处理的去噪和边缘提取,实验结果表明忆阻细胞神经网络在图像处理的应用中具有良好的效果.该文提出的忆阻细胞神经网络可以减小电路尺寸,提高运算速度,电路结构具有更紧凑、更通用的优点,有助于促进人工神经网络的硬件实现.
A new memristor-based cellular neural network is proposed in this paper.We have improved the traditional memristive bridge circuit,besides the advantages of the traditional synaptic bridge circuit,the improved memristive bridge circuit has a more simplified circuit and simplified weight change conditions.At the same time,we have used SPICE circuit simulation to achieve the operation of synaptic weight.In addition,we have applied the memristor-based cellular neural network to image denoising and edge extraction.Experimental results show that the memristor-based cellular neural network has a good effect in the application of image processing.The proposed memristor-based cellular neural network can reduce the size of circuit and improve the speed of operation,and the circuit structure is more compact and universal,which helps to promote the hardware implementation of artificial neural network.
作者
吴洁宁
闫登卫
王丽丹
段书凯
WU Jiening;YAN Dengwei;WANG Lidan;DUAN Shukai(School of Information Science and Technology,Fudan University,Shanghai 200433,China;College of Artificial Intelligence,Southwest University,Chongqing 400715,China)
出处
《西南师范大学学报（自然科学版）》
CAS
2022年第3期1-8,共8页
Journal of Southwest China Normal University(Natural Science Edition)
基金
国家重点研发计划项目(2018YFB1306600)
国家自然科学基金(62076207,62076208,u20A20227)。
关键词
忆阻器
细胞神经网络
突触桥电路
图像处理
memristor
cellular neural/nonlinear network
synaptic bridge circuit
image processing
