Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Detecting Abnormal Health Conditions in Smart Home Using a Drone

Pronob Kumar Barman

TL;DR

The paper tackles the problem of detecting abnormal health conditions, specifically falls, in smart homes using a low-cost drone (Ryze Tello) and vision-based methods. It proposes FallNet, a CNN-based classifier operating on RGB imagery to identify falls, integrated into an end-to-end framework where the drone scans, detects a person, classifies the fall, and triggers emergency alerts. Experimental results on public and in-house datasets show very high performance, with reported validation accuracy reaching 1.0000 after several epochs, and a tested precision of 0.9948 in the abstract, demonstrating feasibility for indoor, autonomous fall monitoring. The study highlights practical benefits for independent living and rapid rescue while acknowledging limitations such as dataset size, drone reliability, camera quality, and the need for autonomous, secure, and scalable enhancements for real-world deployment.

Abstract

Nowadays, detecting aberrant health issues is a difficult process. Falling, especially among the elderly, is a severe concern worldwide. Falls can result in deadly consequences, including unconsciousness, internal bleeding, and often times, death. A practical and optimal, smart approach of detecting falling is currently a concern. The use of vision-based fall monitoring is becoming more common among scientists as it enables senior citizens and those with other health conditions to live independently. For tracking, surveillance, and rescue, unmanned aerial vehicles use video or image segmentation and object detection methods. The Tello drone is equipped with a camera and with this device we determined normal and abnormal behaviors among our participants. The autonomous falling objects are classified using a convolutional neural network (CNN) classifier. The results demonstrate that the systems can identify falling objects with a precision of 0.9948.

Detecting Abnormal Health Conditions in Smart Home Using a Drone

TL;DR

The paper tackles the problem of detecting abnormal health conditions, specifically falls, in smart homes using a low-cost drone (Ryze Tello) and vision-based methods. It proposes FallNet, a CNN-based classifier operating on RGB imagery to identify falls, integrated into an end-to-end framework where the drone scans, detects a person, classifies the fall, and triggers emergency alerts. Experimental results on public and in-house datasets show very high performance, with reported validation accuracy reaching 1.0000 after several epochs, and a tested precision of 0.9948 in the abstract, demonstrating feasibility for indoor, autonomous fall monitoring. The study highlights practical benefits for independent living and rapid rescue while acknowledging limitations such as dataset size, drone reliability, camera quality, and the need for autonomous, secure, and scalable enhancements for real-world deployment.

Abstract

Nowadays, detecting aberrant health issues is a difficult process. Falling, especially among the elderly, is a severe concern worldwide. Falls can result in deadly consequences, including unconsciousness, internal bleeding, and often times, death. A practical and optimal, smart approach of detecting falling is currently a concern. The use of vision-based fall monitoring is becoming more common among scientists as it enables senior citizens and those with other health conditions to live independently. For tracking, surveillance, and rescue, unmanned aerial vehicles use video or image segmentation and object detection methods. The Tello drone is equipped with a camera and with this device we determined normal and abnormal behaviors among our participants. The autonomous falling objects are classified using a convolutional neural network (CNN) classifier. The results demonstrate that the systems can identify falling objects with a precision of 0.9948.
Paper Structure (21 sections, 1 equation, 9 figures, 1 table)

This paper contains 21 sections, 1 equation, 9 figures, 1 table.

Table of Contents

  1. Introduction
  2. Proposed Approach
  3. Related Work
  4. Work Related to UAV Systems
  5. Work Related to Robots
  6. Data Collection
  7. Design of Hardware
  8. Device Setup and Requirements
  9. Study Design
  10. Participant Instructions
  11. Proposed Methodology
  12. Convolutional Neural Network (CNN)
  13. Training and Implementation
  14. Experimental Results
  15. Datasets
  16. ...and 6 more sections

Figures (9)

  • Figure 1: Proposed Approach Flowchart
  • Figure 2: Sample Images of Fall Detection
  • Figure 3: Ryze Tello Drone
  • Figure 4: CNN Architecture a8
  • Figure 5: Overall Architecturea5
  • ...and 4 more figures