Title: An intelligent visual hand motion tracking system for home-based rehabilitation

Authors: Yi Zhang, Yuan Luo, Ling Li

Addresses: Centre of Intelligent System and Robotics, Automation College, Chongqing University of Posts and Telecommunications, Chongqing, 400065, China. ' Centre of Intelligent System and Robotics, Automation College, Chongqing University of Posts and Telecommunications, Chongqing, 400065, China. ' Centre of Intelligent System and Robotics, Automation College, Chongqing University of Posts and Telecommunications, Chongqing, 400065, China

Abstract: Human motion tracking is the key technology for intelligent rehabilitation systems in which the movement of disabled people can be localised and learned and any undesired motion behaviour can be corrected to reach an expectation. This paper introduces a real-time tracking system of human hand motion, specifically used for home based rehabilitation. A hybrid tracking algorithm based on an improved Camshift and a Kalman filter is proposed. In the tracking process, Camshift converts the raw image to a flesh colour probability distribution image via a colour probability table. Considering the hand motion and flesh colour probability distribution, the size and location of a search window are initialised by using the object histogram model. Computed information is used to adjust current search window|s size and location, then the location of the hand are found. Experimental results show that this algorithm is fast, accurate, and robust. It can not only deal with the skin colour interference problems, but also handle complex background.

Keywords: improved Camshift; Kalman filter; colour probability distribution; intelligent rehabilitation; disabled people; disabilities; localisation; motion behaviour; real-time tracking; hand motion; home based rehabilitation; human motion tracking; identification; skin colour interference; complex backgrounds.

DOI: 10.1504/IJMIC.2010.034573

International Journal of Modelling, Identification and Control, 2010 Vol.10 No.3/4, pp.213 - 221

Published online: 10 Aug 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article