MFE-HAR

multiscale feature engineering for human activity recognition using wearable sensors

Jianchao Lu, Xi Zheng, Quan Z. Sheng, Zawar Hussain, Jiaxing Wang, Wanlei Zhou

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

Abstract

Human activity recognition plays a key role in the application areas such as fitness tracking, healthcare and aged care support. However, inaccurate recognition results may cause an adverse effect on users or even an unpredictable accident. In order to improve the accuracy of human activity recognition, multi-device and deep learning based approaches have been proposed. However, they are not practical on a daily basis due to the limitations that devices are difficult to wear, and deep learning requires large training dataset and incurs expensive computational costs. To address this problem, we propose a novel approach, multiscale feature engineering for human activity recognition (MFE-HAR), which exploits the properties of arm movement from global and local scales using the accelerometer and gyroscope sensors on a single wearable device. Our method takes advantage of having important features at multiple scales over previous single-scale methods. We evaluated the performance of the proposed method on two public datasets and achieved the mean classification accuracy of 93% and 98% respectively. Our proposed system performs better than the state of the art multi-device based approaches, and is more practical for real-world applications.
Original languageEnglish
Title of host publicationMobiQuitous '19
Subtitle of host publicationProceedings of the 16th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services
Place of PublicationNew York, NY
PublisherAssociation for Computing Machinery (ACM)
Pages180-189
Number of pages10
ISBN (Electronic)9781450372831
DOIs
Publication statusPublished - 2019
Event16th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services - Houston, United States, Houston, United States
Duration: 12 Nov 201914 Nov 2019
Conference number: 16
http://mobiquitous.org

Conference

Conference16th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services
Abbreviated titleMobiQuitous
CountryUnited States
CityHouston
Period12/11/1914/11/19
Internet address

Keywords

  • Activity recognition
  • pervasive technologies for healthcare
  • mobile and wearable computing systems and services
  • Pervasive technologies for healthcare
  • Mobile and wearable computing systems and services

Fingerprint Dive into the research topics of 'MFE-HAR: multiscale feature engineering for human activity recognition using wearable sensors'. Together they form a unique fingerprint.

  • Cite this

    Lu, J., Zheng, X., Sheng, Q. Z., Hussain, Z., Wang, J., & Zhou, W. (2019). MFE-HAR: multiscale feature engineering for human activity recognition using wearable sensors. In MobiQuitous '19: Proceedings of the 16th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services (pp. 180-189). New York, NY: Association for Computing Machinery (ACM). https://doi.org/10.1145/3360774.3360787