Human gait modeling and gait analysis based on Kinect
- 1 May 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10504729,p. 3173-3178
Abstract
Real-time monitoring of elderly movement can provide valuable information regarding an individual's degree of functional rehabilitation. Many laboratory-based studies have described various gait detection systems with different wearable inertial sensors, but only limited number of papers addressed the issues by using some non-wearable sensors. A practical method of gait information detection and gait analysis is proposed in the paper using an inexpensive Microsoft Kinect fixed on the midpoint of lower extremity rehabilitation robot. The horizontal distances between Kinect plane and every mark pasted on lower extremity are acquired. Taken the characteristics of gait distance series into consideration, the Autoregressive Moving Average (ARMA) model is established to reflect the changing rule of gait status. Combined with the Kalman filter, gait information reflecting rehabilitation status at next moment is predicted accurately. The method regarding the gait detection and gait analysis is verified by amounts of gait experiments finally.Keywords
This publication has 5 references indexed in Scilit:
- Full body gait analysis with Kinect2012
- Wireless slips and falls prediction system2012
- The Elderly Fall Risk Assessment and Prediction Based on Gait Analysis2011
- Minimalistic control of a compass gait robot in rough terrain2009
- Implementation of a Real-Time Human Movement Classifier Using a Triaxial Accelerometer for Ambulatory MonitoringIEEE Transactions on Information Technology in Biomedicine, 2006