In the existing SVM-based fall detection algorithms, the fall actions and the activities of daily living (ADLs) are similar in sample size. In real life, however, there are far more ADLs than fall actions. Thus, the seemingly accurate detection in experiments often does not apply to real life. To solve the problem, this paper takes acceleration and angle as feature vectors, and introduces the asymmetrical support vector machine (SVM) algorithm. The penalty coefficient was configured by changing the diagonal matrix parameters of the kernel function, and the hyperplane was adjusted to approximate the fall action with the smallest possible sample size, seeking to accurately determine the occurrence of fall actions. Through experimental simulation, it is verified that the proposed model can accurately detect 99.2% of fall actions.
fall detection, activities of daily living (ADLs), asymmetrical support vector machine (SVM), acceleration and angle.
This work was supported by the science and technology fund of hebei agricultural university (NO.LG20150202).
1. Korea National Statistical Office, 2010 aging statistics, http://www.index.go.kr/ egams/stts/jsp/ potal/stts/PO_STTS_IdxMain.jsp?idx_cd=1010, August 9, 2013.
2. R. Igual, C. Medrano, I. Plaza, Challenges, issues and trends in fall detection systems, 2013, BioMedical Engineering OnLine, vol. 12, no. 1, p. 66.
3. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Web–based Injury Statistics Query and Reporting System (WISQARS) [online]. Accessed December 30, 2010.
4. C. Rougier, J. Meunier, A. St-Arnaud, J. Rousseau, Robust video surveillance forfall detection based on human shape deformation, 2011, IEEE Trans Circuits Syst for Video Technol, vol. 21, pp. 611-622.
5. M. Mubashir, L. Shao, L. Seed, A survey on fall detection: Principles and approaches, 2012, Neurocomputing, vol. 100, pp. 144-152.
6. D. Nicholas Olivieri, I. Gomez Conde, X.A. Vila Sobrino, Eigenspace-based falldetectionand activity recognition from motion templates and machine learning, 2012, Expert Systems with Applications, vol. 39, no. 5, pp. 5935-5945.
7. C. Zhang, Y. Tian, E. Capezuti, Privacy preserving automatic fall detection for elderly using RGBD cameras, 2012, International Conference on Computers Helping People with Special Needs, Springer-Verlag, Berlin, pp.625-633.
8. Y. Li, K.C. Ho, M. Popescu, A microphone array system for automatic fall detection, 2012, IEEE Transactions on Bionmedical Engineering, vol. 59, no.2, pp.1291-1301.
9. L. Hazelhoff, J. Han, de With PHN,Video-based fall detection in the home using principal component analysis, 2008, Conference on Advanced Concepts for Intelligent Vision Systems, 10th International, Juan-les-Pins, France, pp. 298-309.
10. H. Rimminen, J. Lindström, M. Linnavuo, R. Sepponen, Detection of falls among the elderly by a floor sensorusing the electric near field, 2010, IEEE Trans Inf Technol Biomed, vol. 14, no. 4, pp. 1475-1476.
11. O. Ojetola, E.I. Gaura, J. Brusey, Fall detection with wearable sensors-safe (smart fall detec-tion), 2011, 7th International Conference on Nottingham, 2011, UK, pp. 318-321.
12. D.S. Chen, W. Feng, Yu.Zhang, A wearable wireless fall detection system with accelerators, 2011, Proceedings of the 2011 IEEE International Conference on Robotics and Biomimetics, Karon Beach, Phuket, Thailand, pp. 2259-2263.
13. S. Abbate, M. Avvenuti, F. Bonatesta, G. Cola, P. Corsini, A. Vecchio, A smartphone-based fall detection system, 2012, Pervasive and Mobile Computing, vol. 8, no.6, pp. 883-899.
14. O. Aziz, S.N. Robinovitch, An analysis of the accuracy of wearable sensors for classifying the causes of falls in humans, 2011, IEEE Trans Neural Syst Rehabil, vol. 19, no. 6, pp. 670-676.
15. P. Mostarac, R. Malaric, M. Jurcevic, et al., System for monitoring and fall detection of pa-tients using mobile 3-axis accelerometers sensors, 2011, Conference on Medical Measurements and Applications Proceedings, Bari, Italy, pp. 456-459.
16. H. Kerdegari, K. Samsudin, A.R. Ramli, S. Mokaram, Evaluation of fall detection classification approaches, 2012, In Proceedingsof the 4th International Conference on Intelligent and Advanced Systems, Kuala Lumpur, Malaysia, pp.131-136.
17. T. Zhang, J. Wang, L. Xu, P. Liu, Fall detection by wearable sensor and one-class SVM algorithm, 2006, In Lecture Notesin Control and Information Science, vol. 345, pp. 858-863.
18. C. Doukas, I. Maglogiannis, F. Tragkas, D. Liapis, G. Yovanof, Patient fall detection using support vector machines, 2007, Int Fed Inf Process 2007, vol. 247, pp. 147-156.
19. P. Pierleoni, L. Pernini, A. Belli, L. Palma, SVM-based fall detection method for elderly people using Android low-cost smartphones, 2015,Sensors Applications Symposium (SAS), 2015, Zadar, Croatia, pp. 236-241.
20. S. Shan, T. Yuan, A wearable pre-impact fall detector using feature selection and support vector machine, 2010, In Proceedings of the IEEE 10th International Conference on Signal Processing, Beijing, China, pp. 1686-1689.
21. J. Cheng, X. Chen, M. Shen, A framework for daily activity monitoring and fall detection based on surface electromyography and accelerometer signals, 2013, IEEE J. Biomed and Health Inform., vol. 17, no. 1, pp. 38-45.
22. J. Jacob, T. Nguyen, D.Y.C. Lie., et al., A fall detection study on the sensors placement location and a rule-based multi-thresholds algorithm using both accelerometer and gyroscopes, 2011, IEEE International Conference on Fuzzy Systems, pp. 666-671.