计算机与数字工程2026，Vol.54Issue(1)：13-16,4.DOI:10.3969/j.issn.1672-9722.2026.01.003
基于协调注意力模块的运动想象脑电识别方法
Motor Imagination EEG Recognition Method Based on Coordinated Attention Module
摘要
Abstract
To solve the problem of low recognition rate due to the small number of channels and low signal-to-noise ratio of motor imagery EEG,this paper introduces coordinate attention(CA)into convolutional neural networks and proposes a method of motor imagery EEG recognition based on coordinated attention module.Firstly,the two-dimension time-frequency graph obtained from the pre-processed EEG signal is used as the input of the model,and the motion image intention recognition model is obtained by training the convolutional neural network model.Finally,the method is applied to the data set and a series of comparative experi-ments are carried out with the excellent deep learning methods in recent years.The experimental results show that CA module can be flexibly used in convolutional neural networks,and can improve the accuracy of model recognition without increasing the calculation cost of the model,which indicates that the proposed method is effective in classification recognition.关键词
运动想象脑电信号/卷积注意力模块/协调注意力模块/卷积神经网络Key words
motor imagery-EEG/convolutional attention module/coordinated attention module/convolutional neural net-work分类
数理科学引用本文复制引用
周成诚,曾庆军,杨康,胡家铭,韩春伟..基于协调注意力模块的运动想象脑电识别方法[J].计算机与数字工程,2026,54(1):13-16,4.基金项目
国家自然科学基金项目(编号:11574120) （编号:11574120）
江苏省产业前瞻与共性关键技术项目(编号:BE201803)资助. （编号:BE201803）
