TECH OFFER

AIoT-based Real-time Monitoring and Behavioural Pattern Recognition for Human Activities

KEY INFORMATION

TECHNOLOGY CATEGORY:
Infocomm - Video/Image Processing
Infocomm - Video/Image Analysis & Computer Vision
Show more >
TECHNOLOGY READINESS LEVEL (TRL):
LOCATION:
Singapore
ID NUMBER:
TO174339

TECHNOLOGY OVERVIEW

This technology offer presents an Artificial Intelligence of Things (AIoT) based real-time monitoring system for elderly. The system is based on a hybrid combination of Artificial Intelligence (AI) and existing monitoring devices such as CCTV cameras and pressure sensors to detect various orientations and activities of the elderly autonomously.  

With this integrated solution, it provides real-time monitoring for elderly within a suitable environment through identifying any abnormal activities and behavioral patterns via pattern recognition. Thus, the system aids in alerting caregivers for faster response and intervention if the system detects any abnormal activities from the elderly. This monitoring system can also be customized for different kinds of human activities and environment. 

TECHNOLOGY FEATURES & SPECIFICATIONS

This technology offer is an integration of newly developed AI models and existing monitoring solutions: 

  • Based on configurable Long Short-Term Memory (LSTM) with human pose estimation AI model for elders. 
  • Anonymous human skeletal position tracking from the frame sequence. 
  • Facial Recognition technology to recognize high fall risk elders and motion tracking.  
  • Pressure mat on wheelchairs to detect presence of elders. 

POTENTIAL APPLICATIONS

Potential use-cases of this technology: 

  • Predictive fall detection for elders living in nursing homes or hospitals.
  • Personalized tracking of high fall risk elders to render assistance and prevent falls.
  • Optimized scheduling of elderly activity timings to minimize over-exertion. 
  • Further evaluation by doctors through gathering of data from individual elder activity timings. 

Unique Value Proposition

These are some of the benefits of this technology: 

  • Suitable for various human activities and settings.
  • Scalable for crowded environment such as shopping malls or schools. 
  • Supports live streaming from commercial CCTV cameras via standard protocols such as Real Time Streaming Protocol (RTSP).
  • Infer multiple abnormal behavioral patterns within a single video feed. 
RELATED TECH OFFERS
Solar Energy Management System using Computer Vision
Human Motion Energy Harvester
Automated Environmental Control for Indoor Farming
Fast-Curing and Ready-to-Use Glass Fibre Reinforced Polymer (GFRP)
Passive Radiative Self-Cooling Paint
Face Anti-Spoofing Technology
Asset Tracking Device with Customisable Sensors
High Speed and Sensitive Artificial Olfactory Sensor
AI-Aided Analysis of Capsule Endoscopy Images
AI-Assisted Image Labelling Tool for Large Scale Data Labelling Efficiency