Smartphone-based activity recognition is an emerging field of research that enables a large number of human-centric applications in the u-lifecare domain. Currently, major challenges include the development of real-time position independent and lightweight classifier models to recognize the physical activities inside the smartphone environment. In this paper, we propose a real-time position independent physical activity recognizer that utilizes the embedded accelerometer, ambient light and proximity sensors of smartphone to recognize the physical activities. To validate our model, we implement it in an open source Android platform to recognize six physical activities and performed extensive experiments over 10 subjects. We obtained 88% of class-accuracy and 91.55% F-measures. It is expected that our model would be a practical and realistic solution for physical activity recognition due to its unobtrusive nature and real-time classification of activities.
|Title of host publication||2017 IEEE AFRICON|
|Subtitle of host publication||Science, Technology and Innovation for Africa, AFRICON 2017|
|Editors||Darryn R. Cornish|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|Number of pages||6|
|Publication status||Published - 07 Nov 2017|
|Event||IEEE AFRICON 2017 - Cape Town, South Africa|
Duration: 18 Sep 2017 → 20 Sep 2017
|Name||IEEE AFRICON: Proceedings|
|Conference||IEEE AFRICON 2017|
|Period||18/09/2017 → 20/09/2017|
Bibliographical noteFunding Information:
This research work was supported by the ADEC Award for Research Excellence (A2RE) 2015. This research was also supported by Zayed University RIF funding # R17063.
© 2017 IEEE.
Copyright 2018 Elsevier B.V., All rights reserved.
- Activity Recognition
ASJC Scopus subject areas
- Artificial Intelligence
- Signal Processing
- Human-Computer Interaction
- Computer Science Applications