Title: Systematic review of indoor fall detection systems for the elderly using Kinect

Authors: Amina Ben Haj Khaled; Ali Khalfallah; Med Salim Bouhlel

Addresses: Sciences and Technologies of Image and Telecommunications, Higher Institute of Biotechnology, University of Sfax, Sfax, Tunisia ' Sciences and Technologies of Image and Telecommunications, Higher Institute of Biotechnology, University of Sfax, Sfax, Tunisia ' Sciences and Technologies of Image and Telecommunications, Higher Institute of Biotechnology, University of Sfax, Sfax, Tunisia

Abstract: The fall of the elderly presents a major health problem as it may cause fatal injuries. To improve the life quality of the elderly, researchers have developed several fall detection systems. Several sensors have been used to overcome this problem. So far, Microsoft Kinect has been the most used camera-based sensor for fall detection. This motion detector can interact with computers through gestures and voice commands. In this article, we presented a comprehensive survey of the latest fall detection research using the Kinect sensor. We provide an overview of the main features of the two Kinect versions V1 and V2 and compare their performances. Then, we detailed the method used for the articles selection. We provided a classification of the fall detection techniques to highlight the main differences between them. Finally, we concluded that it is not enough to evaluate a system performance under simulated conditions. It is important to test these approaches on old people who are likely to fall.

Keywords: depth sensor; elderly healthcare; fall detection; Kinect V1; Kinect V2; PRISMA; machine learning.

DOI: 10.1504/IJTMCP.2022.123136

International Journal of Telemedicine and Clinical Practices, 2022 Vol.3 No.4, pp.276 - 301

Received: 18 Jul 2019
Accepted: 16 Sep 2019
Published online: 31 May 2022 *