SCL-Fall: Reliable Fall Detection Using mmWave Radar With Supervised Contrastive Learning

Fall is a severe health threat for elders' health care. While existing systems could achieve promising performance under specific scenarios, the required computing resources are usually not affordable, which is not applicable for real-time detection. In this article, we propose SCL-Fall, a real...

Full description

Saved in:
Bibliographic Details
Published in:IEEE Journal of Selected Areas in Sensors 2024, Vol.1, p.237-248
Main Authors: Li, Wenxuan, Zhang, Dongheng, Li, Yadong, Song, Ruiyuan, Hu, Yang, Sun, Qibin, Chen, Yan
Format: Article
Language:English
Subjects:
Accuracy
Contrastive learning
data augmentation
Fall detection
Feature extraction
Millimeter wave radar
neural network
Neural networks
Radar
Real-time systems
Sensors
wireless sensing
Wireless sensor networks
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Fall is a severe health threat for elders' health care. While existing systems could achieve promising performance under specific scenarios, the required computing resources are usually not affordable, which is not applicable for real-time detection. In this article, we propose SCL-Fall, a real-time fall detection system using millimeter wave signal with supervised contrastive learning, which can achieve impressive accuracy with low computation complexity. Specifically, we first extract the signal variation corresponding to human activity with spatial-temporal processing. We incorporate reweighting and denoising techniques in the signal processing process. To enhance the system performance and robustness, we perform data augmentation by shifting, flipping, extracting, and interpolating the signal. Finally, we design a lightweight convolutional neural network to achieve real-time fall detection. Extensive experimental results demonstrate that the proposed system could achieve state-of-the-art performance with limited computation complexity.
ISSN:2836-2071
2836-2071
DOI:10.1109/JSAS.2024.3481205