Inertial Sensor Based Modelling of Human Activity Classes: Feature Extraction and Multi-sensor Data Fusion using Machine Learning Algorithms

Kostas Giokas (Editor), Laszlo Bokor (Editor), Frank Hopfgartner (Editor), Tahmina Zebin, Patricia J Scully, Krikor B Ozanyan

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    Abstract

    Wearable inertial sensors are currently receiving pronounced interest due to applications in unconstrained daily life settings, ambulatory monitoring and pervasive computing systems. This research focuses on human activity recognition problem, in which inputs are multichannel time series signals acquired from a set of body-worn inertial sensors and outputs are automatically classified human activities. A general-purpose framework has been presented for designing and evaluating activity recognition system with six different activities using machine learning algorithms such as support vector machine (SVM) and artificial neural networks (ANN). Several feature selection methods were explored to make the recognition process faster by experimenting on the features extracted from the accelerometer and gyroscope time series data collected from a number of volunteers. In addition, a detailed discussion is presented to explore how different design parameters, for example, the number of features and data fusion from multiple sensor locations - impact on overall recognition performance.
    Original languageEnglish
    Title of host publicationInstitute for Computer Sciences, Social Informatics and Telecommunications Engineering 2017
    EditorsKostas Giokas, Laszlo Bokor, Frank Hopfgartner
    Place of PublicationCham
    PublisherSpringer Nature
    Pages306-314
    Number of pages9
    Volume181
    ISBN (Electronic)978-3-319-49655-9
    ISBN (Print)978-3-319-49654-2
    DOIs
    Publication statusPublished - 31 Dec 2016

    Publication series

    NameInstitute for Computer Sciences, Social Informatics and Telecommunications Engineering 2017
    Volume181

    Keywords

    • accelerometer data á feature
    • algorithms á human activity
    • data-fusion á machine learning
    • extraction á
    • inertial measurement unit á
    • recognition

    Research Beacons, Institutes and Platforms

    • Manchester Institute for Collaborative Research on Ageing

    Fingerprint

    Dive into the research topics of 'Inertial Sensor Based Modelling of Human Activity Classes: Feature Extraction and Multi-sensor Data Fusion using Machine Learning Algorithms'. Together they form a unique fingerprint.

    Cite this