Research on the Taxonomy of Activity Recognition Based on Inertial Sensors

Abstract:

This paper introduces some techniques on classifying human activities with inertial sensors and point out a number of characteristics of classification algorithm. The goal of human activity recognition is to automatically analyze ongoing activities from people who wear inertial sensor. Initially, we provide introduce information about the activity recognition, such as the way of acquisition, sensors used and the steps of activity recognition using machine learning algorithm. Next, we focus on the classification techniques together with a detailed taxonomy, and the classification techniques implemented and compared in this study are: Decision Tree Algorithm (DTA), Bayesian Decision Making (BDM), Support Vector Machines (SVM), Artificial Neural Networks (ANN) and Hidden Markov Model (HMM)^[. Finally, we make a summarize about it investigate the directions for future research.

• Periodical: Advanced Materials Research (Volume 823)
• Pages: 107-110
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.823.107
• Online since: October 2013
• Authors: Zi Ming Xiao, Yu Long Shi, Yong Xue, Feng Hu, Yu Chuan Wu
• Keywords: Activity Recognition, Classification Techniques, Feature Vectors, Inertial Sensor

References

[1] Kerem Altun, Billur Barshan and Orkun Tuncel, Comparative study on classifying human activities with miniature inertial and magnetic sensors. Pattern Recognition 43 (2010) 3605–3620.

DOI: 10.1016/j.patcog.2010.04.019

[2] Chang Woo HAN and Shin Jae KANG, Implementation of HMM-Based Human Activity recognition Using Single Triaxial Accelerometer, , IEICE trans. fundamentals, vol. E93-A, NO. 7 July (2010).

DOI: 10.1587/transfun.e93.a.1379

[3] S. Paiyarom,P. Tungamchit,R. Keinprasit, and P, Kayasith, Activity monitoring system using dynamic time warping for the elderly and disabled people, Pro. Int. Conf. Computer, Control and Comm- -unication, pp.1-4, Feb. (2009).

DOI: 10.1109/ic4.2009.4909158

[4] S.J. Preece J.Y. Golermas L.P.J. Kenney, and D. Howard, A comparison of feature extraction methods for the classification of dynamic activity from accelerometer data, IEEE Tran, Biomed. Eng., vol. 56, no. 3, pp.871-879, March (2009).

DOI: 10.1109/tbme.2008.2006190

[5] Z.Y. He and L.W. Jin, Activity recognition from acceleration data using AR model representation and SVM, Proc. Inc. Conf. Machine Learning and Cybernetics , vol. 4, pp.2245-2250, July (2008).

DOI: 10.1109/icmlc.2008.4620779

[6] J. R. Quinlan. Induction of Decision Trees. Machine Learning, 1(1): 81–106, (1986).

[7] Yuchuan Wu. Ronghua Chen. Intelligent clothing for automated recognition of human physical activities in free-living environment, The Journal of The Textile Institute 01. Nov. 2011, page 1-11 ISSN0040-5000.

DOI: 10.1080/00405000.2011.611641

[8] Yousef Ajami Alotaibi. Comparative Study of ANN and HMM to Arabic Digits Recognition Systems, JKAU: Eng. Sci., Vol. 19 No. 1, pp: 43-60(2008A.D. /1429 A.H. ).

DOI: 10.4197/eng.19-1.3

[9] S. Haykin, Neural Networks: A Comprehensive Foundation, Macmillan Publishing, New York, USA, (1994).

[10] M.T. Hagan, H.B. Demuth, M.H. Beale, Neural Network Design, PWS, Boston, MA, USA, (1996).

[11] H. -K. Lee J.H. Kim, An HMM-based threshold model approach for gesture recogniton , , IEEE Tran. Pattern Anal. Mach. Intell., vol. 21, no. 10, pp.961-973, Oct. (1999).

DOI: 10.1109/34.799904