Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ElderFallGuard: Real-Time IoT and Computer Vision-Based Fall Detection System for Elderly Safety

Tasrifur Riahi, Md. Azizul Hakim Bappy, Md. Mehedi Islam

TL;DR

ElderFallGuard addresses the critical problem of reliable, real-time fall detection for the elderly using a non-invasive vision-based approach. It builds a pipeline around MediaPipe Pose for 33 landmark extraction, a custom 12-pose dataset of 7200 samples, and a Random Forest classifier, augmented with a temporal logic that requires Pose6 persistence and motion reduction before confirming a fall. The system integrates an intelligent Telegram-based alerting mechanism that sends caregiver notifications with a snapshot and includes cooldown to prevent alert fatigue. The approach demonstrates perfect metrics on the collected test set and offers a practical, end-to-end solution with potential for real-world impact, pending validation in diverse environments and privacy considerations.

Abstract

For the elderly population, falls pose a serious and increasing risk of serious injury and loss of independence. In order to overcome this difficulty, we present ElderFallGuard: A Computer Vision Based IoT Solution for Elderly Fall Detection and Notification, a cutting-edge, non-invasive system intended for quick caregiver alerts and real-time fall detection. Our approach leverages the power of computer vision, utilizing MediaPipe for accurate human pose estimation from standard video streams. We developed a custom dataset comprising 7200 samples across 12 distinct human poses to train and evaluate various machine learning classifiers, with Random Forest ultimately selected for its superior performance. ElderFallGuard employs a specific detection logic, identifying a fall when a designated prone pose ("Pose6") is held for over 3 seconds coupled with a significant drop in motion detected for more than 2 seconds. Upon confirmation, the system instantly dispatches an alert, including a snapshot of the event, to a designated Telegram group via a custom bot, incorporating cooldown logic to prevent notification overload. Rigorous testing on our dataset demonstrated exceptional results, achieving 100% accuracy, precision, recall, and F1-score. ElderFallGuard offers a promising, vision-based IoT solution to enhance elderly safety and provide peace of mind for caregivers through intelligent, timely alerts.

ElderFallGuard: Real-Time IoT and Computer Vision-Based Fall Detection System for Elderly Safety

TL;DR

ElderFallGuard addresses the critical problem of reliable, real-time fall detection for the elderly using a non-invasive vision-based approach. It builds a pipeline around MediaPipe Pose for 33 landmark extraction, a custom 12-pose dataset of 7200 samples, and a Random Forest classifier, augmented with a temporal logic that requires Pose6 persistence and motion reduction before confirming a fall. The system integrates an intelligent Telegram-based alerting mechanism that sends caregiver notifications with a snapshot and includes cooldown to prevent alert fatigue. The approach demonstrates perfect metrics on the collected test set and offers a practical, end-to-end solution with potential for real-world impact, pending validation in diverse environments and privacy considerations.

Abstract

For the elderly population, falls pose a serious and increasing risk of serious injury and loss of independence. In order to overcome this difficulty, we present ElderFallGuard: A Computer Vision Based IoT Solution for Elderly Fall Detection and Notification, a cutting-edge, non-invasive system intended for quick caregiver alerts and real-time fall detection. Our approach leverages the power of computer vision, utilizing MediaPipe for accurate human pose estimation from standard video streams. We developed a custom dataset comprising 7200 samples across 12 distinct human poses to train and evaluate various machine learning classifiers, with Random Forest ultimately selected for its superior performance. ElderFallGuard employs a specific detection logic, identifying a fall when a designated prone pose ("Pose6") is held for over 3 seconds coupled with a significant drop in motion detected for more than 2 seconds. Upon confirmation, the system instantly dispatches an alert, including a snapshot of the event, to a designated Telegram group via a custom bot, incorporating cooldown logic to prevent notification overload. Rigorous testing on our dataset demonstrated exceptional results, achieving 100% accuracy, precision, recall, and F1-score. ElderFallGuard offers a promising, vision-based IoT solution to enhance elderly safety and provide peace of mind for caregivers through intelligent, timely alerts.
Paper Structure (20 sections, 5 figures, 1 table)

This paper contains 20 sections, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background and Motivation
  3. Objective
  4. Literature Review
  5. System Design & Methodology
  6. Data Acquisition: The Foundation
  7. Real-Time Pose Estimation with MediaPipe
  8. Feature Extraction for Classification
  9. Fall Pose Classification
  10. Temporal Monitoring Logic for Fall Confirmation
  11. Intelligent Alerting System via Telegram
  12. Experiments
  13. Experimental Setup
  14. Evaluation Metrics
  15. Results
  16. ...and 5 more sections

Figures (5)

  • Figure 1: Body parts injured in non-fatal falls by age groups in Bangladesh Wadhwaniya2017
  • Figure 2: System Architecture
  • Figure 3: MediaPipe Landmark Detection
  • Figure 4: Telegram alert screenshot showing message and image
  • Figure 5: Confusion Matrix of (a) Random Forest, (b) K-Nearest Neighbors, (c) Support Vector Machine and (d) Gradient Boosting.