International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
Call for Papers | Fully Refereed | Open Access | Double Blind Peer Reviewed

ISSN: 2319-7064

Downloads: 118 | Views: 185

Research Paper | Electronics & Communication Engineering | India | Volume 6 Issue 1, January 2017

Fall Detection for Elderly People in Indoor Environment using Kinect Sensor

Disha Pathak | V. K. Bhosale [3]

Abstract: Health problems of the elderly people are becoming more and more severe with the growth of aging population. The accidents like falling down are one of the major risks for the older adults living alone at home and need to be paid more attention. Through the analysis of the existing fall detection techniques, a more easy and quick algorithm about human body fall detection based on the human skeleton extraction using Kinect sensor is proposed in this paper. This algorithm consists of two parts, which are moving target depth of image acquisition, processing of depth image and identification of human skeleton. The realization of the detection algorithm is based on the tracked key joints acquired from the Kinect sensor. Two parameters are extracted, by comparing these values with the threshold, the system judges whether human falls down. Once a fall event is detected an alert message is sent to a predefined number using SIM900A GSM modem. Using new technologies such as the Kinect sensor with matlab could bring innovative ways to build smart systems that could use to observe the elderly people at homes and notify the caretakers in case of falling events Based on a real dataset of 50 people, experimental results indicate that the proposed method can realize human's fall detection with 94.65% accuracy in indoor environment.

Keywords: Kinect sensor, Fall detection, Depth image, Skeleton extraction

Edition: Volume 6 Issue 1, January 2017,

Pages: 1956 - 1960

How to Download this Article?

Type Your Valid Email Address below to Receive the Article PDF Link

Verification Code will appear in 2 Seconds ... Wait