Completed

Human Behavior Classification

Real-time multi-patient behavior detection using radar and deep CNNs

Human Behavior Classification

Overview

To address potential gaps in patient monitoring in hospitals, a novel patient behavior detection system using mmWave radar and deep convolutional neural network (CNN) is proposed. The system supports the simultaneous recognition of multiple patients' behaviors in real-time, addressing critical needs in healthcare facilities. The system was tested for real-time operation and obtained very good inference accuracy when predicting each patient's behavior in multi-patient scenarios, providing a privacy-preserving alternative to camera-based monitoring systems.

Key Highlights

  • Healthcare Monitoring
  • Multi-Patient Detection
  • Real-time Processing
  • Privacy Preservation
  • Deep CNN

Methodology

Deep CNN-based behavior classification from radar micro-Doppler signatures. Real-time processing for simultaneous multi-patient monitoring.

Technologies

mmWave Radar Deep CNN Micro-Doppler Analysis Real-time Inference Multi-Target Processing

Project Details

Start Date
September 2017
End Date
April 2019
Status
Completed

Resources

Live Demo

Related Publications

Real-Time Human Motion Behavior Detection via CNN using mmWave Radar

R. Zhang, S. Cao

IEEE Sensors Letters, 2019

View Paper
Multiple Patients Behavior Detection in Real-Time using mmWave Radar and Deep CNNs

F. Jin, R. Zhang, A. Sengupta, S. Cao, S. Hariri, N. K. Agarwal, S. K. Agarwal

IEEE Radar Conference, 2019

View Paper
Back to All Projects