基于改进深度森林的运动想象脑电分类方法研究
作者:
作者单位:

天津工业大学

作者简介:

通讯作者:

中图分类号:

TH79 TN911.7 R318

基金项目:

国家自然科学基金项目(62071328)天津市技术创新引导专项基金(21YDTPJC00540,21YDTPJC00550).


Classification method of motor imagery EEG based on improved deep forest
Author:
Affiliation:

Tiangong University

Fund Project:

The National Natural Science Foundation of China (62071328) Tianjin Technology Innovation Guidance Project (Fund) of China (21YDTPJC00540,21YDTPJC00550)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    基于运动想象的脑电信号是用户在执行不同运动想象任务时采集到的不同脑区的电信号.受到用户的大脑结构和头皮状态等因素影响,采集到的运动想象任务信号之间混乱,从而导致大量信号被错分.提出了一种基于改进深度森林的运动想象任务信号分类方法.该方法利用变长粒子群算法强大的寻优能力,为深度森林中每一层的随机森林和完全随机森林预测的类概率值搜寻最优权重,并将此权重赋予对应的类概率值,以此实现对结果修正目的.利用BCI竞赛IV的数据集2a评估所提出方法的有效性.研究结果表明:相比传统的深度森林,该方法对四分类运动想象脑电信号实现了更高的分类准确率.所提方法根据分类器预测的结果进行学习,对提升分类器性能的研究具有重要意义.

    Abstract:

    EEG signals based on motor imagery are electrical signals acquired from different brain regions when subjects perform different motor imagery tasks. Influenced by factors such as the difference of brain structures and scalp states of different subjects, the acquired signals of different motor imagery tasks are usually confused. This will result in the misclassifications for a large number of motor imagery signals. A new classification algorithm for motor imagery tasks based on improved deep forest was proposed in this study. The powerful optimization ability of the variable-length particle swarm algorithm was adopted to search for the optimal weights for probability values of each class predicted by the random forest and completely random forest in the deep forest. The weights were assigned to the corresponding class probability values, so as to realize the purpose of the result correction. The data set 2a of BCI Competition IV was adopted to evaluate the efficacy of the proposed algorithm. The results show that the proposed algorithm achieves a higher classification accuracy for the four classes of motor imagery signals in comparison with the traditional deep forest. The algorithm investigated in this study can learn from the predicted results of the classifier, which is of great significance for the improvement of the classifier performance.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-01-18
  • 最后修改日期:2021-12-08
  • 录用日期:2021-12-09
  • 在线发布日期: 2022-01-02
  • 出版日期: