bullet Sensors & Transducers Journal

    (ISSN 1726- 5479)

 

205.767

2008 e-Impact Factor

25 Top Downloaded Articles

Journal Subscription 2010

Editorial Calendar 2009

Submit an Article

Editorial Board

Current Issue

Sensors & Transducers journal's cover

Sensors & Transducers Journal 2008

Sensors & Transducers Journal 2007

2000-2002 S&T e-Digest Contents

2003 S&T e-Digest Contents

2004 S&T e-Digest Contents

2005 S&T e-Digest Contents

2006 S&T e-Digest Contents

 

Best Articles 2008

 

 

 

MEMS: From Micro Devices to Wireless Systems

Vol. 7, Special Issue, October 2009, pp.213-223

 

 

Bullet

 

Classifying Transition Behaviour in Postural Activity Monitoring

 

James BRUSEY, Ramona REDNIC and Elena GAURA

Coventry University, Priory St, Coventry, CV1 5FB, UK

Tel.: +44 2476887688

E-mail: j.brusey@coventry.ac.uk

 

 

Received: 28 August 2009   /Accepted: 28 September 2009   /Published: 12 October 2009

 

Abstract: A few accelerometers positioned on different parts of the body can be used to accurately classify steady state behaviour, such as walking, running, or sitting. Such systems are usually built using supervised learning approaches. Transitions between postures are, however, difficult to deal with using posture classification systems proposed to date, since there is no label set for intermediary postures and also the exact point at which the transition occurs can sometimes be hard to pinpoint. The usual bypass when using supervised learning to train such systems is to discard a section of the dataset around each transition. This leads to poorer classification performance when the systems are deployed out of the laboratory and used on-line, particularly if the regimes monitored involve fast paced activity changes. Time-based filtering that takes advantage of sequential patterns is a potential mechanism to improve posture classification accuracy in such real-life applications. Also, such filtering should reduce the number of event messages needed to be sent across a wireless network to track posture remotely, hence extending the system’s life. To support time-based filtering, understanding transitions, which are the major event generators in a classification system, is a key. This work examines three approaches to post-process the output of a posture classifier using time-based filtering: a naïve voting scheme, an exponentially weighted voting scheme, and a Bayes filter. Best performance is obtained from the exponentially weighted voting scheme although it is suspected that a more sophisticated treatment of the Bayes filter might yield better results.

 

Keywords: Posture classification, Evaluation of performance for posture classification instrumentation, Dealing with postural transitions, Data annotation, transitions filtering algorithms and experimental results, Context: case study of bomb disposal missions operatives monitoring

 

Acrobat reader logo Click <here> or title of paper to download the full pages article (2.26 Mb)

 

 

 

 


1999 - 2009 Copyright ©, International Frequency Sensor Association (IFSA). All Rights Reserved.


Home - News - Links - Archives - Tools - Standardization - Patents - Marketplace - Projects - Wish List - Subscribe - Search - Membership - Submit Press Release

 Members Area -Sensors Portal -Training Courses - S&T Digest - For advertisers - Bookstore - Forums - Polls - Sensor Jobs - e-Shop - Site Map