4 October 2018

Research pick: Smartphone activity - "A two-phase human activity classification design using accelerometer data from smartphone"

Your smartphone could be used to monitor your level of activity – whether you are running, walking or standing, according to research in China. The benefits might be in healthcare, checking up on patients with mobility issues of fall risk as well as in health and fitness apps that allow the user to set targets for different levels of activity.

The team has developed a two-stage method for analyzing the data from the smart phone’s built-in accelerometer. The team’s algorithm looks at the acceleration in three dimensions, x, y, and z and can then determine whether the person carrying the phone is running, walking, or standing with an accuracy of more than 97%. The next steps will include improving accuracy still further as well as developing the technology so that it can distinguish other physical actions, perhaps sporting activity such as cycling or rowing.

Ali, S.A. and Amin, R.U. (2018) ‘A two-phase human activity classification design using accelerometer data from smartphone’, Int. J. Intelligent Systems Technologies and Applications, Vol. 17, No. 3, pp.281–291.

No comments: