Multi-sensor Based Human Balance Analysis
The human balance ability is investigated using a multi-sensor human balance assessment system. Based on the pressure sensor, the gyroscope, the accelerometer and the magnetometer, the quantitative perception of human balance under different postures is realized. The characteristics of human balance ability are extracted through time domain methods and a new hybrid feature extraction. The results demonstrate that the hybrid feature extraction method with the support vector machine method can effectively classify and evaluate the human balance ability under different postures.
KeywordsHuman balance Multi-sensor Hybrid feature extraction Support Vector Machine
- 2.Barton, J.E., Roy, A., Sorkin, J.D., et al.: An engineering model of human balance control—part I: biomechanical model. J. Biomech. Eng. 138(1), 1–11 (2016)Google Scholar
- 3.Terekhov, Y.: Stabilometry as a diagnostic tool in clinical medicine. Can. Med. Assoc. J. 115(7), 631–633 (1976)Google Scholar
- 9.Wang, D.Q., Wang, F.F., Zhu, W.C., et al.: Prediction for non-stationary non-linear time series based on empirical mode decomposition. Syst. Eng. 32(5), 138–142 (2014)Google Scholar
- 11.Liu, K., Wang, H., Xiao, J.Z., et al.: Analysis of human standing balance by largest Lyapunov exponent. Comput. Intell. Neurosci. 2015(2), 151–154 (2015)Google Scholar