Self-Adaptive Multi-Sensor Activity Recognition Systems Based on Gaussian Mixture Models
Journal Title: Informatics - Year 2018, Vol 5, Issue 3
Abstract
Personal wearables such as smartphones or smartwatches are increasingly utilized in everyday life. Frequently, activity recognition is performed on these devices to estimate the current user status and trigger automated actions according to the user’s needs. In this article, we focus on the creation of a self-adaptive activity recognition system based on inertial measurement units that includes new sensors during runtime. Starting with a classifier based on Gaussian Mixture Models, the density model is adapted to new sensor data fully autonomously by issuing the marginalization property of normal distributions. To create a classifier from that, label inference is done, either based on the initial classifier or based on the training data. For evaluation, we used more than 10 h of annotated activity data from the publicly available PAMAP2 benchmark dataset. Using the data, we showed the feasibility of our approach and performed 9720 experiments, to get resilient numbers. One approach performed reasonably well, leading to a system improvement on average, with an increase in the F-score of 0.0053, while the other one shows clear drawbacks due to a high loss of information during label inference. Furthermore, a comparison with state of the art techniques shows the necessity for further experiments in this area.
Authors and Affiliations
Martin Jänicke, Bernhard Sick and Sven Tomforde
Keywords
Related Articles
- EP ID EP44145
- DOI https://doi.org/10.3390/informatics5030038
- Views 243
- Downloads 0
Real-Time and Embedded Detection of Hand Gestures with an IMU-Based Glove†
This article focuses on the use of data gloves for human-computer interaction concepts, where external sensors cannot always fully observe the user’s hand. A good concept hereby allows to intuitively switch the interac...
Player–Game Interaction and Cognitive Gameplay: A Taxonomic Framework for the Core Mechanic of Videogames
Cognitive gameplay—the cognitive dimension of a player’s experience—emerges from the interaction between a player and a game. While its design requires careful consideration, cognitive gameplay can be designed only ind...
Hybrid Design Tools—Image Quality Assessment of a Digitally Augmented Blackboard Integrated System
In the last two decades, Interactive White Boards (IWBs) have been widely available as a pedagogic tool. The usability of these boards for teaching disciplines where complex drawings are needed, we consider debatable i...
Ambient Assisted Living for Improvement of Health and Quality of Life—A Special Issue of the Journal of Informatics
The demographic change with respect to the ageing of the population has been a worldwide trend. As a direct result, it has been recognised as causing substantial social transformation in the 21st century [1]. By 2050,...
Towards Clustering of Mobile and Smartwatch Accelerometer Data for Physical Activity Recognition
Mobile and wearable devices now have a greater capability of sensing human activity ubiquitously and unobtrusively through advancements in miniaturization and sensing abilities. However, outstanding issues remain around...