Integrating Features for Accelerometer-based Activity Recognition

Activity recognition is the problem of predicting the current action of a person through the motion sensors worn on the body. The problem is usually approached as a supervised classification task where a discriminative model is learned from known samples and a new query is assigned to a known activity label using learned model. The challenging issue here is how to feed this classifier with a fixed number of features where the real input is a raw signal of varying length. In this study, we consider three possible feature sets, namely time-domain, frequency domain and wavelet-domain statistics, and their combinations to represent motion signal obtained from accelerometer reads worn in chest through a mobile phone. In addition to a systematic comparison of these feature sets, we also provide a comprehensive evaluation of some preprocessing steps such as filtering and feature selection. The results determine that feeding a random forest classifier with an ensemble selection of most relevant time-domain and frequency-domain features extracted from raw data can provide the highest accuracy in a real dataset.

[1]  In-Cheol Kim,et al.  Human Activity Recognition as Time-Series Analysis , 2015 .

[2]  Long Tang,et al.  Instantaneous Real-Time Kinematic Decimeter-Level Positioning with BeiDou Triple-Frequency Signals over Medium Baselines , 2015, Sensors.

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

[4]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[5]  Pedro Larrañaga,et al.  A review of feature selection techniques in bioinformatics , 2007, Bioinform..

[6]  T Togawa,et al.  Classification of Acceleration Waveforms during Walking by Wavelet Transform , 1997, Methods of Information in Medicine.

[7]  I. Maqsood,et al.  Random Forests and Decision Trees , 2012 .

[8]  Sohail Asghar,et al.  A REVIEW OF FEATURE SELECTION TECHNIQUES IN STRUCTURE LEARNING , 2013 .

[9]  David Howard,et al.  A Comparison of Feature Extraction Methods for the Classification of Dynamic Activities From Accelerometer Data , 2009, IEEE Transactions on Biomedical Engineering.

[10]  Lukun Wang,et al.  Recognition of Human Activities Using Continuous Autoencoders with Wearable Sensors , 2016, Sensors.

[11]  Silvestro Micera,et al.  Developing the next generation of hybrid neuroprosthetic systems. , 2009, IEEE transactions on bio-medical engineering.

[12]  Ahmed Ghoneim,et al.  A Triaxial Accelerometer-Based Human Activity Recognition via EEMD-Based Features and Game-Theory-Based Feature Selection , 2016, IEEE Sensors Journal.

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

[14]  Petia Radeva,et al.  Human Activity Recognition from Accelerometer Data Using a Wearable Device , 2011, IbPRIA.

[15]  Diogo R. Ferreira,et al.  Preprocessing techniques for context recognition from accelerometer data , 2010, Personal and Ubiquitous Computing.

[16]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..