Scholarlink

With the rapid development of intelligent manufacturing, there is an increasing demand for fatigue monitoring among workers in production processes. Although various fatigue monitoring methods have been proposed, most are either highly invasive or pose privacy concerns. To address these issues, this study introduces a non-invasive, contactless fatigue monitoring method using an intelligent vibration-sensing chair. The proposed method integrates an improved Convolutional Neural Network (CNN) with a Long Short-Term Memory (LSTM) network for fatigue state assessment. Experimental results demonstrate that the CNN-LSTM model effectively identifies the fatigue states of non-manual workers, achieving a classification accuracy of 99.0%. This research provides a user-friendly, privacy-preserving, and efficient solution for monitoring mental fatigue among non-manual laborers.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Non-Invasive Evaluation of Fatigue State of Non-Manual Workers Based on CNN-LSTM Network

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Zhen Ma,
  • He Xu,
  • Jielong Dou,
  • Yi Qin

摘要

With the rapid development of intelligent manufacturing, there is an increasing demand for fatigue monitoring among workers in production processes. Although various fatigue monitoring methods have been proposed, most are either highly invasive or pose privacy concerns. To address these issues, this study introduces a non-invasive, contactless fatigue monitoring method using an intelligent vibration-sensing chair. The proposed method integrates an improved Convolutional Neural Network (CNN) with a Long Short-Term Memory (LSTM) network for fatigue state assessment. Experimental results demonstrate that the CNN-LSTM model effectively identifies the fatigue states of non-manual workers, achieving a classification accuracy of 99.0%. This research provides a user-friendly, privacy-preserving, and efficient solution for monitoring mental fatigue among non-manual laborers.