Abstract:Effective fall detection is a critical component of indoor health monitoring for individuals. The most widely used wearable devices are not convenient, especially for the elderly. Another approach of passive solutions usually use RGB camera, millimeter-wave radar, or WiFi. However, they bring about issues of privacy infringement and/or electromagnetic interference. Among them, visible light-based systems can avoid these issues, however, they require customized light sources and highly sensitive receivers, making it expensive for deployment. Motivated by these previous works, we aim to design a passive, low-invasiveness, and low-cost fall detection system. In this work, we propose LiFall, a passive fall detection system based on illumination and visible light communication (VLC) networks, which eliminates the need for hardware modifications to current devices. LiFall detects human falls by analyzing the intensity variation of the received optical signals captured by the low-cost photodetectors. Subsequently, we propose an innovative signal segmentation algorithm tailored for accurately segmenting human activities in the time series. Compared to existing double-window segmentation schemes, our method enhances the signal integrity by 20%. In order to identify falls, we analyze the temporal statistical features and waveform characteristics derived from human activity signals and adopt the lightweight classifiers of support vector machines (SVM) and random forests (RF) for fall detection. We deploy LiFall system in various typical indoor scenarios and collect human fall and other daily activity data for system performance evaluation. Experimental results demonstrate that LiFall is able to achieve a fall detection accuracy exceeding 90% in different typical indoor settings without compromising illumination and communication.

Fall Detection with Unobtrusive Infrared Array Sensors

Robust Unobtrusive Fall Detection Using Infrared Array Sensors.

Fall Detection System Based on Infrared Array Sensor and Multi-Dimensional Feature Fusion

Fall Detection and Personnel Tracking System Using Infrared Array Sensors

A Real-time Fall Detection System Using ToF Depth Images.

Unobtrusive High-Generalization Fall Detection: A Domain-Generalization Framework

An Unobtrusive Fall Detection System Using Ceiling-mounted Ultra-wideband Radar

FallViewer: A Fine-Grained Indoor Fall Detection System with Ubiquitous Wi-Fi Devices.

Real-time fall detection system based on deep learning and infrared array sensors

Applying deep learning technology for automatic fall detection using mobile sensors

An Integrated Sensing and Communication System for Fall Detection and Recognition Using Ultra-Wideband Signals

ISFDNet: Using Low-Resolution Infrared Array Sensors to Detect Elderly Fall in Home Elevator Scenarios

A Class-Imbalanced Deep Learning Fall Detection Algorithm Using Wearable Sensors

A Method for Fast Fall Detection

A Self-Learning Fall Detection System for Elderly Persons Using Depth Camera

RF-Based Fall Monitoring Using Convolutional Neural Networks

LiFall: Passive Indoor Fall Detection System Based on Illumination and Visible Light Communication Networks

Fall Detection Method for Infrared Videos Based on Spatial-Temporal Graph Convolutional Network

Fall Detection for Elderly People Using Machine Learning

Fall Detection Algorithm Based on Inertial Sensor and Hierarchical Decision

Fall Detection With UWB Radars and CNN-LSTM Architecture