Convolutional Neural Networks for Human Activity Recognition Using Body-Worn Sensors
Journal Title: Informatics - Year 2018, Vol 5, Issue 2
Abstract
Human activity recognition (HAR) is a classification task for recognizing human movements. Methods of HAR are of great interest as they have become tools for measuring occurrences and durations of human actions, which are the basis of smart assistive technologies and manual processes analysis. Recently, deep neural networks have been deployed for HAR in the context of activities of daily living using multichannel time-series. These time-series are acquired from body-worn devices, which are composed of different types of sensors. The deep architectures process these measurements for finding basic and complex features in human corporal movements, and for classifying them into a set of human actions. As the devices are worn at different parts of the human body, we propose a novel deep neural network for HAR. This network handles sequence measurements from different body-worn devices separately. An evaluation of the architecture is performed on three datasets, the Oportunity, Pamap2, and an industrial dataset, outperforming the state-of-the-art. In addition, different network configurations will also be evaluated. We find that applying convolutions per sensor channel and per body-worn device improves the capabilities of convolutional neural network (CNNs).
Authors and Affiliations
Fernando Moya Rueda, René Grzeszick, Gernot A. Fink, Sascha Feldhorst and Michael Ten Hompel
Keywords
Related Articles
- EP ID EP44134
- DOI https://doi.org/10.3390/informatics5020026
- Views 295
- Downloads 0
Multimodal Interaction of Contextual and Non-Contextual Sound and Haptics in Virtual Simulations
Touch plays a fundamental role in our daily interactions, allowing us to interact with and perceive objects and their spatial properties. Despite its importance in the real-world, touch is often ignored in virtual envi...
The Effects of Motion Artifacts on Self-Avatar Agency
One way of achieving self-agency in virtual environments is by using a motion capture system and retargeting user’s motion to the virtual avatar. In this study, we investigated whether the self-agency is affected when mo...
Multi-Gateway-Based Energy Holes Avoidance Routing Protocol for WSN
In wireless sensor networks (WSNs), efficient energy conservation is required to prolong the lifetime of the network. In this work, we have given emphasis on balanced energy consumption and energy holes avoidance. This...
Detecting Transitions in Manual Tasks from Wearables: An Unsupervised Labeling Approach†
Authoring protocols for manual tasks such as following recipes, manufacturing processes or laboratory experiments requires significant effort. This paper presents a system that estimates individual procedure transition...
Modelling Digital Knowledge Transfer: Nurse Supervisors Transforming Learning at Point of Care to Advance Nursing Practice
Limited adoption of mobile technology for informal learning and continuing professional development within Australian healthcare environments has been explained primarily as an issue of insufficient digital and ehealth...