Title :
Research on key algorithm of human motion trac~ing for intelligent rehabilitation system
Author :
Li, Ling ; Li, Yanle ; Luo, Yuan
Author_Institution :
Dept. of Phys. & Electron. Inf., Hengshui Univ., Hengshui, China
Abstract :
This paper focuses on the technology of human motion tracking based on video. First, It discusses the Camshift (Continuously Adaptive Meanshift) tracking algorithm. Camshift tracking algorithm is a tracking method based on the object´s color character, and is widely used for the parameter-less and fast calculation. Then, on the basis of keeping the merit of Camshift and improving its disadvantages, this paper proposes an improved hand motion tracking algorithm-A video hand motion tracking algorithm based on hybrid Camshift tracking algorithm and Kalman filter motion prediction. This paper designs a video hand motion tracking system and does a lot of simulation experiments, and the experiments´ results are well examined. The simulation results demonstrate that the video hand tracking algorithm based on hybrid Camshift tracking algorithm and Kalman filter motion prediction can do automatic motion detection and tracking with high reliability. It can overcome the skin color interference problem, and has obvious robustness and practicability.
Keywords :
Kalman filters; image colour analysis; motion estimation; object tracking; patient rehabilitation; video signal processing; Kalman filter motion prediction; automatic motion detection; continuously adaptive meanshift tracking algorithm; human motion tracking; hybrid camshift tracking algorithm; intelligent rehabilitation system; object color character; skin color interference problem; video hand motion tracking algorithm; Decision support systems; World Wide Web; Camshift; Kalman filter; color probability distribution; hand tracking; rehabilitation and search window;
Conference_Titel :
Control and Decision Conference (CCDC), 2012 24th Chinese
Conference_Location :
Taiyuan
Print_ISBN :
978-1-4577-2073-4
DOI :
10.1109/CCDC.2012.6244205