Utilizing SmartPhones for Non-Invasive Activity Recognition

Daniel Kelly and Brian Caulfield


Intelligent Data Analysis, Pattern Recognition, Activity Recognition


In order for activity recognition systems to be implemented in real world scenarios such as health and wellness monitoring, the activity sensing modality must unobtrusively fit the human environment rather than forcing humans to adhere to sensor specific conditions. In this paper, we propose an a set of techniques which allow us to carry out automatic activity recognition using only very minor sensor placement conditions. Modern smart phones represent a ubiquitous computing device which has gone through mainstream adoption and we therefore focus on activity recognition techniques which can adapt to the most common on-body phone placement locations. We describe a set of orientation independent features, a set of techniques which identifies the location of the sensor and an activity classification framework which automatically adapts to the detected location.

Important Links:

Go Back