Online Recognition with Postural Transition Awareness - Smartphone-Based Human Activity Recognition - page 124

activities. Furthermore, the repeated detection of this unknown events can be also an
indication that the system is not working correctly. For example, if the sensor is not
located in the right position or if the activities performed by a new user seem not to
be recognized properly, etc. This idea should be further studied and it is proposed as
a future direction of research.
This work explored probability estimates as a measure of how likely an activity
is to be carried out. This has shown to be useful for performing filtering techniques
when it is also combined with the idea of interrelationship between different activ-
ities. Henceforth, it would be interesting to explore novel approaches to make the
recognition of activities more robust by applying probabilistic approaches such as
Markov chains on smartphones. These can, for example, define the PTs as transitions
between states (BAs) and use activity probability estimates as the observations or
input to the system.
References
A. Bulling, U. Blanke, B. Schiele, A tutorial on human activity recognition using body-worn inertial
sensors. ACM Comput. Surv. 46 , 33 (2014)
A.M. Khan, Y.-K. Lee, S.Y. Lee, T.-S. Kim, A triaxial accelerometer-based physical-activity recog-
nition via augmented-signal features and a hierarchical recognizer. IEEE Trans. Inf. Technol.
Biomed. 14 , 1166-1172 (2010)
O. Lara, M. Labrador, A survey on human activity recognition using wearable sensors. IEEE Com-
mun. Surv. Tutorials 1 , 1-18 (2012)
B. Najafi, K. Aminian, F. Loew, Y. Blanc, P.A. Robert, Measurement of stand-sit and sit-stand
transitions using a miniature gyroscope and its application in fall risk evaluation in the elderly.
IEEE Trans. Biomed. Eng. 49 , 843-851 (2002)
R. Rednic, E. Gaura, J. Kemp, J. Brusey, in Fielded autonomous posture classification systems:
design and realistic evaluation , ACIS International Conference on Software Engineering, Artifi-
cial Intelligence, Networking and Parallel/Distributed Computing (2013)
S. Zhang, P. McCullagh, C. Nugent, H. Zheng, in Activity monitoring using a smart phone's
accelerometer with hierarchical classification , International Conference on Intelligent Environ-
ments (2010)