Scholarlink

This article presents a fall detection method designed for elderly individuals, primarily based on using a specialized detection network called YOLO and a tracking algorithm to monitor individuals. A dataset of falls involving elderly people was collected, and the model was trained and fine-tuned on this dataset to optimize performance. The DeepSORT algorithm is employed for tracking, and finally, an implementation integrating IoT is proposed. The model achieved strong performance on the dataset, with a precision of 0.996 for fall detection, a recall of 0.9, and effective detection capabilities on real videos.

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

Fall Tracking and Detection Using YOLO on IoT Devices

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Zakaria Benhaili,
  • Farah Oubelkas,
  • Youssef Balouki,
  • Lahcen Moumoun

摘要

This article presents a fall detection method designed for elderly individuals, primarily based on using a specialized detection network called YOLO and a tracking algorithm to monitor individuals. A dataset of falls involving elderly people was collected, and the model was trained and fine-tuned on this dataset to optimize performance. The DeepSORT algorithm is employed for tracking, and finally, an implementation integrating IoT is proposed. The model achieved strong performance on the dataset, with a precision of 0.996 for fall detection, a recall of 0.9, and effective detection capabilities on real videos.