Towards a Real-Time Fall Detection System using Kinect Sensor

2. Related Work

The demand for monitoring systems, especially for fall detection, has increased within the healthcare industry with the rapid growth of the population of the elderly in the world. It has become very important to develop intelligent monitoring systems, especially vision-based systems, which can automatically monitor and detect falls. It has been proved that the medical consequences of a fall are highly contingent upon the response and rescue time (Mubashir et al 2012).

The goal aimed by researchers is to realize a monitoring system that is effectively able to distinguish between an activity of daily living (ADL) and a fall even with complex indoor scenarios that include occlusions and obstacles. Several methods have been proposed and the challenge is to detects fall in real-time. Many works have been devoted for elderly persons fall detection. (Igual et al.2013), Webster and Celik (2014), Mubashir et al. (2012) and Noury et al. (2008) reviewed the principal methods used in existing fall detection approaches. We can classify the different approaches according to the used equipment into two categories: Wearable device based and camera (vision) based. Detailed surveys can be found in Igual et al. (2013), Webster and Celik, (2014) and Mubashir et al. (2012). In the following, we summarize the different methods.