Determination of the most effective human body regions for activity recognition

Year 2018, Volume: 20 Issue: 2, 372 - 381, 01.12.2018

Gökmen Aşçıoğlu , Yavuz Şenol

https://doi.org/10.25092/baunfbed.487066

Abstract

Monitoring daily activities and providing feedback from life activities performed can prevent many diseases and improve the quality of life of individuals. In the academic studies carried out, it was evaluated that the data obtained from a single sensor placed on the chest was used to define the resultant activity using various complex algorithms. In this study, the most effective body regions were identified for activity identification. For this purpose, total of four accelerometers were placed in the chest, shoulder, limb and arm regions. Data sets were collected for different activities including walking, running, jumping, and sit-to-stand. The performances of artificial neural networks were examined using single or multi-sensor data sets for activity recognition. The results show that using multi-sensor in effective parts has more positive impact on neural network performance.

Keywords

Daily activities, classification, artificial neural networks, accelerometer

Aktivite tanımlama için en etkin vücut bölgelerinin belirlenmesi

Abstract

Günlük aktivitelerin izlenmesi ve gerçekleştirilen yaşam aktivitelerinden geri bildirim sağlanması birçok hastalığı önleyebilir ve bireylerin yaşam kalitesini yükseltir. Gerçekleştirilen akademik çalışmalarda genellikle göğüs üzerine yerleştirilen tek bir sensörden elde edilen verilerin çeşitli kompleks algoritmalarla kullanılması sonucu aktivite tanımlaması yapıldığı değerlendirilmiştir. Bu çalışmada ise aktivite tanımlama için en efektif vücut bölgeleri belirlenmiştir. Bu amaçla, toplamda dört ivme sensörü göğüs, omuz, bacak ve kol bölgelerine yerleştirilmiştir. Yürüme, koşma, zıplama ve oturma-kalkma aktiviteleri süresince veriler toplandı. Tekli ve çoklu sensör verileri ile kullanılmasının aktivite tanımlama için yapay sinir ağları performansına etkisi incelenmiştir. Sonuçlar etkin bölgelerde çoklu sayıda sensör kullanmanın performansa daha olumlu yansıdığını ortaya çıkarmıştır.

Keywords

Günlük aktiviteler, sınıflandırma, yapay sinir ağları, ivme sensörü

References

• World Health Organization. Prevalence of insufficient physical activity. http://www.who.int/(Erişim Tarihi: 14.04.2018).
• Wahid, A. vd., Quantifying the Association Between Physical Activity and Cardiovascular Disease and Diabetes: A Systematic Review and Meta-Analysis, Journal of the American Heart Association, 5(9), (2016).
• Kyu, H.H. vd., Physical activity and risk of breast cancer, colon cancer, diabetes, ischemic heart disease, and ischemic stroke events: systematic review and dose-response meta-analysis for the Global Burden of Disease Study, The BMJ, (2016).
• Brymer, E., Davids, K., Designing environments to enhance physical and psychological benefits of physical activity: a multidisciplinary perspective. Sports Medicine, 46(7), 925-926, (2016).
• Schuldhaus, D., Leutheuser, H., Eskofier, B.M., Classification of daily life activities by decision level fusion of inertial sensor data, In Proceedings of the 8th International Conference on Body Area Networks, ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, 77-82, (2013).
• Khan, A.M., Lee, Y.K., Lee, S.Y., & Kim, T.S., A triaxial accelerometer-based physical-activity recognition via augmented-signal features and a hierarchical recognizer, IEEE Transactions on Information Technology in Biomedicine, 14(5), 1166-1172, (2010).
• Lara, O.D., Pérez, A.J., Labrador, M.A., Posada, J.D., Centinela: A human activity recognition system based on acceleration and vital sign data, Pervasive and Mobile Computing, 8(5), 717-729, (2012).
• Dadashi, F., Arami, A., Crettenand, F., Millet, G.P., Komar, J., Seifert, L., Aminian, K., A hidden markov model of the breaststroke swimming temporal phases using wearable inertial measurement units, In Body Sensor Networks (BSN), 2013 IEEE International Conference on, 1-6, (2013).
• Balli, S., Sağbaş, E.A., Akıllı saat algılayıcıları ile insan hareketlerinin sınıflandırılması, Süleyman Demirel Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 21(3), 1-11, (2017).
• El Achkar, C.M., Massé, F., Arami, A., Aminian, K., Physical activity recognition via minimal in-shoes force sensor configuration. In Proceedings of the 7th International Conference on Pervasive Computing Technologies for Healthcare. ICST (Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering), 256-259 (2013).
• Hagan, M., Demuth, H.B., Jess, O.D. and Beale, M., Neural Network Design, , USA, (2014).
• Haykin, S., Neural Networks: A Comprehensive Foundation, Pearson Education, 9, India, (2005).