Serious injuries associated with falls in the elder age group is a big health care problem. Some studies have estimated that one in every three adults aging above 65 will fall at least once per year. Elderly fall can lead to serious injuries such as traumatic brain injuries, bone fractures leading to immobility. According to ACC’s statistical report, there are approximately 168 000 active claims with a total cost of $170 million of injuries associated with elderly falls in the year 2015-2016. This number has increased consistently over the past decade and is expected to keep increasing in the future due to the ageing population.
Fall detection systems are used to detect and alert when a fall event has occurred. There are various types of fall detection system on the market. However, they can be divided into 3 types including wearable device based, ambience sensor based and camera vision based.
In this project, our aim is to design and implement a camera (vision) based fall detection system using computer vision techniques to automatically detect falls and notify relevant people. The system uses Microsoft Kinect’s RGB camera, IR camera and IR projector to collect data in real time. The system manipulates and analyses the collected data using a set of computer vision techniques and machine learning algorithms. The project outcome will include a mobile application and a cloud-based server which included the data processor. The mobile application will be simple to use and will be used as one of the user end to receive the alerts from the back-end servers when falls occur.
Undergraduate
•
A Scalable software architecture consists of a mobile application and cloud-based servers
• Implement a fall detecting algorithm to dissect and accurately analyse information from the motion sensors
• Elaborate deep learning algorithm and incorporate it into the detection algorithm
None
Lab allocations have not been finalised