摘要
在脑机接口的研究中,针对运动想象脑电信号的特征识别,提出了一种基于概率协作表示的分类方法(probabilistic collaborative representation based classification,ProCRC),通过比较测试样本在每个类别中的最大可能性,从而确定其所属的类别。采用BCI竞赛数据库Ⅲ中的数据集Ⅰ,利用S变换进行特征提取,然后对不同的分类器进行比较,以分类准确率作为评价标准验证了该算法的有效性。该算法的分类准确率能够达到90%,为脑机接口系统分类算法的研究提供了新思路。
In the research of brain-computer interface,a classification method for recognizing the features of motor imagery EEG signals based on probabilistic collaborative representation（Pro CRC） was proposed in this paper.The maximum likelihood that a test sample belonged to each of the multiple classes was compared,so as to determine the final classification that it belonged to.Performance of this method was tested using the data set of BCI competition Ⅲ.Firstly,the S transform was used to extract the electroencephalography features,and then different classifiers were compared.Finally,the classification accuracy was used as the evaluation criterion to verify the effectiveness of the algorithm.The accuracy of the algorithm proposed in this paper could reach 90%,which provided a new idea for the research of the classification algorithm of the brain-computer interface system.
作者
崔丽霞
杨济民
常洪丽
CUI Li-xia;YANG Ji-min;CHANG Hong-li(School of Physics and Electronics, Shandong Normal University, Jinan 250358, Chin)
出处
《山东科学》
CAS
2018年第2期105-112,共8页
Shandong Science
基金
山东省重点研发计划(2017GGX10102)
关键词
脑机接口
概率协作表示
S变换
运动想象
brain-computer interface
probabilistic collaborative representation
S-transform
motor imagery
