Loading...

Design and Implementation of a GAIT Analysis System Using Kinect for Clinical Application

Jamali Soosefi, Zahra | 2014

464 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46730 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Behzadipour, Saeed
  7. Abstract:
  8. To date various commercial systems were used in the gait analysis area. These systems have some difficulties for clinical use, such as being indwell, making trouble in movement and high prices. The Kinect sensor does not have problems of these systems. If the error of sensor is acceptable, Kinect sensor is a suitable choice for application in clinics. The possibility of utilization of the Kinect sensor as a gait analysis system has been studied in this research. The sensor errors in calculation of gait parameters such as lower limb joints angle, stride time, stride length and spatial coordinates of joints were computed. In previous researches the Kinect sensor error has been calculated for upper limb joints angle and the subject was standing motionless in front of the sensor. The Kinect sensor has also been used to record some identified motions on behalf of chiropractors. In researches which the Kinect sensor has been used for gait analysis, limited parameters of gait has been calculated and there has not been any full study about the position of the sensor toward the motion path. The purpose of this research is determining the Kinect sensor errors during walking, according to effective parameters on the sensors accuracy such as the type of joint and the sensors position. The error calculation method used in this research was placing the Kinect sensor in different positions toward the motion path. In each position the data was recorded by both Kinect sensor and commercial motion capture system. After data recording and noise filtering the angle of ankle, knee and hip joints, stride time and length and spatial coordinates of joints were calculated by the output of both systems. In calculation of Kinect error the obtained values from the commercial motion capture system were considered as the true value. The results demonstrate that the cutoff frequency 4.5 Hz is the best frequency for noise elimination from the outputs. According to the results, the best position for the Kinect sensor is in front of the subject. When the Kinect sensor was placed in front of the subject, the error of stride time and stride length was obtained 1.4 percent and 4.2 cm respectively. The root mean square error of knee and hip joints angle in this position was computed 4.5 and 4.4 degree respectively. The error of ankle joint angle was higher than the acceptable amount for clinical applications. According to the small length of the foot link, existence of error in foot coordinates measurement causes great error in calculation of ankle joint angle. The results prove the competency of the Kinect sensor as a gait analysis system in clinics
  9. Keywords:
  10. Fault Model ; Kinect Sensor ; Gait Analysis ; Gait Parameters

 Digital Object List

 Bookmark

No TOC