University of Twente Student Theses
Real time posture tracking using Breathline : posture , body position and movement classification using an Arduino, accelerometer and k-Nearest Neighbour classification.
Poot, Martijn (2020) Real time posture tracking using Breathline : posture , body position and movement classification using an Arduino, accelerometer and k-Nearest Neighbour classification.
Full text not available from this repository.
| Full Text Status: | Access to this publication is restricted |
| Embargo date: | 1 August 2022 |
| Abstract: | Breathing is perhaps the most underappreciated function of the human body and mind, it influences not only if we are alive or dead but also our state of mind. Breathing techniques have been practiced by individuals in practices such as Tai-Chi, Qigong and Yoga throughout history. Abdominal breathing is seen as the most natural way of breathing and in order to train this the Breathline wearable was developed. The Breathline wearable makes use of Respiratory Inductance Plethysmography (RIP) to measure breathing and give the user feedback on their performance. Because it is unknown what the influence of posture, body position and movement is on abdominal breathing the clients of this project wanted to know if it is possible to integrate posture classification into the Breathline wearable. In order to solve this the following research question was formulated: ”How can a wearable mounted in the abdominal region be used to detect and classify body position, posture and movement data?”. Background research showed that breathing is influenced by body position and verified the need to integrate posture classification in the Breathline wearable, further research on available sensors concluded that IMU’s are the most viable solution with accelerometers as a close second. A system mounted in the abdominal region making use of an Arduino and an accelerometer is then envisioned and various methods for classification are explored. A Hi-Fi prototype is built making use of k-Nearest Neighbour classification and user tests are performed to validate the accuracy of the prototype as well as to explore the influence of posture and body position on breathing. The prototype system created performs with an accuracy of >90% when classifying posture, body position and movement. No correlations between different postures and respiratory rate and amplitude were found but this could be because of the small sample size, further more in depth research is recommended to find conclusive results. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 50 technical science in general, 53 electrotechnology, 54 computer science |
| Programme: | Creative Technology BSc (50447) |
| Link to this item: | https://purl.utwente.nl/essays/82073 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page