The system is applicable for Factories (Manufacturing industry) and Nursing Home (Healthcare industry).

  • From health care point of view, the instant treatment for injuries of fallen people is very critical for not worsen the injuries.
  • The elderly are subject to higher risks of accidental falls and also higher rate of long-term trauma that affects the quality of life later.
  • Therefore, in developed countries, there is a tremendous demand for supportive products and technologies to improve the safety for the elderly living at Nursing Home or living on their own.
  • The first system to mention is Personal Emergency Response System which composes of a small radio transmitter, a console connecting to the user's telephone with ‚ÄúHELP‚Äù button on it, and an emergency response center that monitors such types of calls.
  • However, the conventional PERS tends to be impractical since the elderly may easily forget to carry the ‚ÄúHELP‚Äù button or they might experience unconscious states of mind as well as physical pain and can‚Äôt press "HELP" button immediately.
  • Timely notification for fall incidence and treatment.
  • Reduce risks of fatality.
  • Computer vision
  • Analytics
  • We developed a solution called Intelligent Emergency Response System (iERS) that is capable of providing automatic sensing of emergencies due to accidental falls of the elderly. The system is also useful in factories to improve labor safety.
  • The system uses the most promising and practical technical solution which is Vision-based technology, rather than Wearable devices and Ambient devices technologies. It is not only to detect fall events accurately but also preserve the privacy and facilitate the freedom of the users.
  • In this project, we develop "Multiview 3D spatial feature - based approaches to fall detection.

Method 1: Fall Detection with Two Cameras based on Heights and Occupied Areas of people

Method 2: Fall Detection based on Human-Ground Contact Areas Usual states of a human, i.e, standing, sitting, lying and in suspicious are significantly different in 3D. Then we can classify human states based on their 3D Spatial Features and analyze State Transitions to identify Fall Event.