WorldCIST'14 - The 2014 World Conference on Information Systems and Technologies

Full Program »

A Simple Movement Classification System for Smartphones with Accelerometer

The penetration rate of smartphones has been growing during the past years. Today's smartphones provide access to the Internet, GPS navigation and are equipped with cameras and various sensors: accelerometer, gyroscope, proximity sensor and light sensor among others. The main objective of this paper is to propose a numeric/linguistic interface whose main characteristic is to obtain the numeric acceleration values along the three axes of the accelerometer and the subsequent conversion to one limited set of linguistic terms. The resulting simple classification of movements is sufficient to classify correctly (from a person’s point of view) the smartphone movements as well as their intensity. The validation tests and the proof of concept presented in this article open the path to the development of applications for physiotherapy and mobile health, especially those aimed at improving health and welfare through motivation for physical activity.


Micael Costa    
Universidade da Beira Interior (UBI)

Paulo Fazendeiro    
Instituto de telecomunicações (IT)


Powered by OpenConf®
Copyright ©2002-2013 Zakon Group LLC