基于蝙蝠算法-随机森林分类算法和脉搏波的精神疲劳检测算法
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TN911.4

基金项目:

国家自然科学基金资助项目(61976030);斯沃德股份有限公司资助项目(2018Q131)


Pulse wave mental fatigue detection algorithm based on BA-RFC
Author:
Affiliation:

Fund Project:

  • 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:

    针对精神疲劳易诱发慢性病和癌症但却难以精准量化评估的问题,提出了一种基于BA-RFC的光电容积脉搏波信号(Photoplethysmographic,PPG)人体精神疲劳检测方法。模拟脑力工作者的疲劳状况,以被试者脉搏波信号为元数据,以蝙蝠算法(BatAlgorithm,BA)和随机森林分类算法(Random forest classification,RFC)相结合,建立疲劳检测模型。首先使用蝙蝠算法对随机森林分类算法中的决策树和分类特性等参数进行优化,通过迭代选取最优的决策树树数和分类特征数,然后根据最优决策树和分类特征进行疲劳识别。仿真结果表明,BA-RFC算法能筛选出与疲劳相关性较高的特征,利用筛选后的特征进行疲劳识别,该方法优化了参数,降低了维度灾难,提高了疲劳检测精度。识别准确率从93.3%提高到96.7%。

    Abstract:

    Mental fatigue is prone to induce chronic diseases and cancer; however, it is difficult to be accurately detected or evaluated. To overcome this problem, we propose mental-fatigue detecting method based on Photoplethysmography (PPG). The fatigue status of mental workers is simulated, their pulsation data acquired and served as metadata. Then, we establish a fatigue-detection model by combining Bat Algorithm (BA) and Random Forest Classification (RFC). In the model, we use BA to optimize the characteristics, such as the number of decision trees and classification. Subsequently, we apply the afore-obtained optimal decision trees and classification features into RFC algorithm to identify the fatigue status. The simulation results show that BA-RFC combined algorithm can screen out the features that are highly correlated with fatigue state, and improve the fatigue recognition accuracy from 93.3% to 96.7%.

    参考文献
    相似文献
    引证文献
引用本文

杨如民,雷倩,余成波. 基于蝙蝠算法-随机森林分类算法和脉搏波的精神疲劳检测算法[J]. 科学技术与工程, 2022, 22(4): 1495-1501.
Yang Rumin, Lei Qian, Yu Chengbo. Pulse wave mental fatigue detection algorithm based on BA-RFC[J]. Science Technology and Engineering,2022,22(4):1495-1501.

复制
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:2021-05-19
  • 最后修改日期:2021-11-15
  • 录用日期:2021-09-29
  • 在线发布日期: 2022-01-28
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注