Author:

Abstract:

More than thirty percent of persons over 65 years fall at least once a year and are often not able to get up again unaided [1][2]. The lack of timely aid can lead to severe complications such as dehydration, pressure ulcers and death. A camera-based fall detection system can provide a solution. For this, several new algorithms have been proposed in the literature recently [3-6]. However, these algorithms are evaluated almost exclusively on data captured in controlled environments, under optimal conditions (simple scenes, perfect illumination and setup of cameras), and with falls simulated by actors. In contrast, we collected a dataset based on real life data, recorded at the place of residence of three older persons over several months. We showed that this poses a significantly harder challenge than the “artificial” datasets used in previous studies. The image quality is typically low. Falls are rare and vary a lot both in speed and nature. We investigated the variation in environment parameters and context during the fall incidents. We found that various complicating factors, such as moving furniture or the use of walking aids, are very common yet almost unaddressed in the literature. Under such circumstances and given the large variability of the data in combination with the limited number of examples available to train the system, we posit that simple yet robust methods incorporating, where available, domain knowledge (e.g. the fact that the background is static or that a fall usually involves a downward motion) seem to be most promising. Based on these observations, we propose a new fall detection system. It is based on background subtraction and simple measures extracted from the dominant foreground object such as aspect ratio, fall angle and head speed. All tests are executed using real life data, which has been recorded at the home of 3 elderly, containing 24 falls. Experiments indicate that a fall detector based on a combination of aspect ratio, head speed and fall angle is preferred. The preliminary detector still has a substantial false alarm rate with a precision of 0.26(±0.07) and a promising recall of 0.9(±0.2). We discuss these results, with special emphasis on particular difficulties encountered under real world circumstances. More details can be found in [7].