Activity recognition of the elderly

The development of context aware services is one proposed way to support independent living for the elderly. Performing test scenarios with the elderly helps when developing the context aware services. However, rigorous testing is not always desirable when working with elderly subjects. Our research proposes to capture the activity data of the subjects to use with a virtual environment and virtual human to test the services. This paper begins a larger set of research by describing a process in which the daily activities of the elderly are captured using accelerometer sensors. The process consists of pre-investigation, data capturing and data postprocessing. Using common activity recognition methods daily activities chosen by two elderly subjects themselves are recognized reasonably well. This allows using the described process in larger experiments to acquire more activity data.

[1]  Michael L. Littman,et al.  Activity Recognition from Accelerometer Data , 2005, AAAI.

[2]  Bernt Schiele,et al.  Analyzing features for activity recognition , 2005, sOc-EUSAI '05.

[3]  Futoshi Naya,et al.  B-Pack: a Bluetooth-based wearable sensing device for nursing activity recognition , 2006, 2006 1st International Symposium on Wireless Pervasive Computing.

[4]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques with Java implementations , 2002, SGMD.

[5]  Tatsuo Nakajima,et al.  Feature Selection and Activity Recognition from Wearable Sensors , 2006, UCS.

[6]  A. Pinz,et al.  Inertial tracking for mobile augmented reality , 2002, IMTC/2002. Proceedings of the 19th IEEE Instrumentation and Measurement Technology Conference (IEEE Cat. No.00CH37276).

[7]  Ingo Mierswa,et al.  YALE: rapid prototyping for complex data mining tasks , 2006, KDD '06.

[8]  Margaret S. Geroch.  Motion capture for the rest of us , 2004 .

[9]  Ling Bao,et al.  Activity Recognition from User-Annotated Acceleration Data , 2004, Pervasive.