Privacy Preserving Automatic Fall Detection for Elderly Using RGBD Cameras

Abstract

In this paper, we propose a new privacy preserving automatic fall detection method to facilitate the independence of older adults living in the community, reduce risks, and enhance the quality of life at home activities of daily living (ADLs) by using RGBD cameras. Our method can recognize 5 activities including standing, fall from standing, fall from chair, sit on chair, and sit on floor. The main analysis is based on the 3D depth information due to the advantages of handling illumination changes and identity protection. If the monitored person is out of the range of a 3D camera, RGB video is employed to continue the activity monitoring. Furthermore, we design a hierarchy classification schema to robustly recognize 5 activities. Experimental results on our database collected under conditions with normal lighting, without lighting, out of depth range demonstrate the effectiveness of the proposal method.