Significant Features for Human Activity Recognition Using Tri-Axial Accelerometers

Bennasar, Mohamed; Price, Blaine; Gooch, Daniel; Bandara, Arosha and Nuseibeh, Bashar (2022). Significant Features for Human Activity Recognition Using Tri-Axial Accelerometers. Sensors, 22(19), article no. 7482.

DOI: https://doi.org/10.3390/s22197482

Abstract

Activity recognition using wearable sensors has become essential for a variety of applications. Tri-axial accelerometers are the most widely used sensor for activity recognition. Although various features have been used to capture patterns and classify the accelerometer signals to recognise activities, there is no consensus on the best features to choose. Reducing the number of features can reduce the computational cost and complexity and enhance the performance of the classifiers. This paper identifies the signal features that have significant discriminative power between different human activities. It also investigates the effect of sensor placement location, the sampling frequency, and activity complexity on the selected features. A comprehensive list of 193 signal features has been extracted from accelerometer signals of four publicly available datasets, including features that have never been used before for activity recognition. Feature significance was measured using the Joint Mutual Information Maximisation (JMIM) method. Common significant features among all the datasets were identified. The results show that the sensor placement location does not significantly affect recognition performance, nor does it affect the significant sub-set of features. The results also showed that with high sampling frequency, features related to signal repeatability and regularity show high discriminative power.

Viewing alternatives

Metrics

Item Actions

Export

You can export this page using these formats Digital Object Identifier - DOI

About

• Item ORO ID
• 85313
• Item Type
• Journal Item
• ISSN
• 1424-8220
• Project Funding Details

• Funded Project Name	Project ID	Funding Body
  STRETCH: Socio-Technical Resilience for Enhancing Targeted Community Healthcare	EP/P01013X/1	EPSRC Engineering and Physical Sciences Research Council
  SAUSE: Secure, Adaptive, Usable Software Engineering	EP/R013144/1	EPSRC (Engineering and Physical Sciences Research Council)
  SERVICE	EP/V027263/1	EPSRC (Engineering and Physical Sciences Research Council)
  Science Foundation Ireland	SFI 13/RC/2094	Not Set

• Keywords
• tri-axial accelerometer; human activity recognition; feature selection; activity of daily living; Physical Activity; classification
• Academic Unit or School
• Faculty of Science, Technology, Engineering and Mathematics (STEM) > Computing and Communications
  Faculty of Science, Technology, Engineering and Mathematics (STEM)
• Copyright Holders
• © 2022 The Authors
• Depositing User
• ORO Import