Scholarlink

Elderly falls mostly refer to critical incidents, which imply a critical need for detection systems that can accurately detect falls. This paper presents an IoT-based fall-detection and activity monitoring system by the integration of supervised machine learning algorithms with MPU6050 motion sensors. The performance of six machine learning algorithms has been tested on collected motion data to detect and classify activities such as sitting, standing, walking, and falling. Results showed the Random Forest model yielded the highest accuracy (96%) and F1-score (95.5%) for detecting falls compared to SVM and outperformed it in classification performance as well as robustness to data variations. The proposed system is guaranteed to be a reliable, user-friendly solution in elderly care due to its real-time functionality and low false alarm rates.

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

Integrating IoT and Machine Learning for Human Fall Detection and Activity Monitoring

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Vijaylaxmi Bittal,
  • Mayur Jain,
  • Shruti Patil,
  • Harshada Patil,
  • Soham Chaudhari

摘要

Elderly falls mostly refer to critical incidents, which imply a critical need for detection systems that can accurately detect falls. This paper presents an IoT-based fall-detection and activity monitoring system by the integration of supervised machine learning algorithms with MPU6050 motion sensors. The performance of six machine learning algorithms has been tested on collected motion data to detect and classify activities such as sitting, standing, walking, and falling. Results showed the Random Forest model yielded the highest accuracy (96%) and F1-score (95.5%) for detecting falls compared to SVM and outperformed it in classification performance as well as robustness to data variations. The proposed system is guaranteed to be a reliable, user-friendly solution in elderly care due to its real-time functionality and low false alarm rates.