Fall Detection for Elder People Using Single Inertial Sensor

DOI

10.2991/iiicec-15.2015.272How to use a DOI?

Keywords

Fall Detection; Behavior Recognition; SVM; Inertial Sensors

Abstract

This paper presents a fall detection system for elder people using wearable single inertial sensor. In contrast with multiple wearable sensors, we deploy only one 3-axis accelerator on human body for continuously monitoring fall incident. Support Vector Machine is exploited to be the classifier for predicting the behavior. The dedicated features including intensity of acceleration, ascending coefficient and descending coefficient are selected for training model. The experimental results have shown that the positive detection rate can reach 92.5% after optimizing SVM parameters.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Wei Zhuang
AU  - Xiang Sun
AU  - Dong Dai
PY  - 2015/03
DA  - 2015/03
TI  - Fall Detection for Elder People Using Single Inertial Sensor
BT  - Proceedings of the 2015 International Industrial Informatics and Computer Engineering Conference
PB  - Atlantis Press
SP  - 1232
EP  - 1235
SN  - 2352-538X
UR  - https://doi.org/10.2991/iiicec-15.2015.272
DO  - 10.2991/iiicec-15.2015.272
ID  - Zhuang2015/03
ER  -