Abstract
Inertial sensors are widely used in the field of human activity recognition (HAR), since this source of information is the most informative time series among non-visual datasets. HAR researchers are actively exploring other approaches and different sources of signals to improve the performance of HAR systems. In this study, we investigate the impact of combining bio-signals with a dataset acquired from inertial sensors on recognizing human daily activities. To achieve this aim, we used the PPG-DaLiA dataset consisting of 3D-accelerometer (3D-ACC), electrocardiogram (ECG), photoplethysmogram (PPG) signals acquired from 15 individuals while performing daily activities. We extracted hand-crafted time and frequency domain features, then, we applied a correlation-based feature selection approach to reduce the feature-set dimensionality. After introducing early fusion scenarios, we trained and tested random forest models with subject-dependent and subject-independent setups. Our results indicate that combining features extracted from the 3D-ACC signal with the ECG signal improves the classifier’s performance F1-scores by and (from to , and to ) for subject-dependent and subject-independent approaches, respectively.
Highlights
With the recent increase in the use of smart phones and wearable devices, we can record and access a plethora of raw data and information from built-in inexpensive sensors.Human activity recognition refers to analyzing these data to extract meaningful information about human daily habits and physical activity patterns [1]
We use hand-crafted features to evaluate the performance of classifiers for human activity recognition (HAR); We investigate the significance of bio-signals and compare the usefulness of ECG and PPG signals in HAR; We investigate the impact of combining a 3D-ACC signal with an ECG signal in recognizing some specific activities in detail
We focus on the 3D-ACC, which is a source of information frequently used in HAR research and applications
Summary
With the recent increase in the use of smart phones and wearable devices, we can record and access a plethora of raw data and information from built-in inexpensive sensors.Human activity recognition refers to analyzing these data to extract meaningful information about human daily habits and physical activity patterns [1]. With the recent increase in the use of smart phones and wearable devices, we can record and access a plethora of raw data and information from built-in inexpensive sensors. One of the most frequently used sources of data for activity recognition purposes is the inertial sensor [3,4,5]. The 3D-ACC sensor outperforms gyroscope and magnetometer, combining 3D-ACC with gyroscope yields better performance in classifying activities [6]. This suggests sensor combination has the potential to offer better classification power for a HAR system
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
