Design of an Activity Recognition System using Smartphone Accelerometer


KIPS Transactions on Software and Data Engineering, Vol. 2, No. 1, pp. 49-54, Jan. 2013
10.3745/KTSDE.2013.2.1.49,   PDF Download:

Abstract

Activity recognition using smartphone accelerometer suffers from the user dependency problem that acceleration patters of one user differ from those of others for the same activity. Moreover, it also suffers from the position dependency problem since a smartphone may be placed in any pockets or hands. In order to overcome these problems, this paper proposes an effective activity recognition method which is less dependent with both specific users and specific positions of the smartphone. Based on the proposed method, we implement a real-time activity recognition system working on an Android smartphone. Throughout some experiments with 6642 examples collected from different users and different positions, we investigate the performance of our activity recognition system.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
I. C. Kim, J. H. Kim, S. H. Nam, S. K. Heo, "Design of an Activity Recognition System using Smartphone Accelerometer," KIPS Transactions on Software and Data Engineering, vol. 2, no. 1, pp. 49-54, 2013. DOI: 10.3745/KTSDE.2013.2.1.49.

[ACM Style]
In Cheol Kim, Joo Hee Kim, Sang Ha Nam, and Se Kyeong Heo. 2013. Design of an Activity Recognition System using Smartphone Accelerometer. KIPS Transactions on Software and Data Engineering, 2, 1, (2013), 49-54. DOI: 10.3745/KTSDE.2013.2.1.49.