DocumentCode
1284918
Title
Image measurement of body position during side-approach transfer motion
Author
Tanimoto, Yuta ; Nanba, Kuniharu ; Tokuhiro, Akihiro ; Ukida, Hideyuki ; Yamamoto, Hiroshi
Author_Institution
Dept. of Med. Eng., Kibikogen Rehabilitation Centre for Employment Injuries, Okayama, Japan
Volume
5
Issue
5
fYear
2011
fDate
8/1/2011 12:00:00 AM
Firstpage
402
Lastpage
409
Abstract
In order to clarify the side-transfer ability of spinal cord injury (SCI) patients, it is necessary to analyse the side-transfer motion of SCI patients. This study describes a method for easy detection of the body position during the transfer motion using a simple human body model and side- and front-view images taken by using two charge-coupled device (CCD) cameras. The proposed simple human body model consists of five cuboids, nine columns and two elliptical cylinders. First, the wheelchair size was measured at the 11 points on its right side using a side-view image. Next, the wheelchair position was detected from the centre position of the right back wheel, and the gap angle between the platform and the wheelchair was detected from the edge position of both back wheels on the side- and front-view images. The positions of the 11 parts of the body model were estimated by matching each part to the side- and front-view images, and then the gravity centre of the body was calculated. Finally, the two images and the centre of gravity calculated every 1 s during the transfer motion were used to realise the movement of the body during the transfer motion.
Keywords
CCD image sensors; edge detection; medical image processing; CCD cameras; SCI patients; body position detection; centre of gravity calculation; charge coupled device cameras; edge position detection; elliptical cylinders; gap angle; human body model; image measurement; side-transfer motion; spinal cord injury patients; time 1 s; wheelchair position;
fLanguage
English
Journal_Title
Image Processing, IET
Publisher
iet
ISSN
1751-9659
Type
jour
DOI
10.1049/iet-ipr.2009.0248
Filename
5963779
Link To Document