基于Camshift和Kalman滤波的仿人机器人手势跟踪
基于Camshift和Kalman滤波的仿人机器人手势跟踪彭娟春， 顾立忠， 苏剑波(上海交通大学自动化系，上海200030)摘要：对仿人机器人MIR-1的双目视觉系统实现实时手势跟踪．通过颜色直方图反投影，将每帧RGB输入图像转换为二维的肤色概率分布图像，基于Camshift算法计算手势跟踪窗口的位置和大小，并用Kalman滤波预测手心位置，有效地解决了背景中大面积肤色干扰和手势部分被遮挡等问题．在仿人机器人MIR一1上完成的手势跟踪实验，验证了此方法的实用性和有效性．关键词：手势跟踪；Camshift；Kalman滤波；颜色概率分布
中图分类号：TP 242．6 文献标识码：A
The Hand Tracking fOr HumanOld RObOt Using Camshift AlgOrithm and Kalman FilterPENG Juan-chun, GU Li-zhong, SU Jian-bo(Dept．of Automation，Shanghai Jiaotong Univ．，Shanghai 200030，China)Abstract：Based on Camshift(Continuously Adaptive Mean Shift)algorithm and the constant—velocity Kal一man filter algorithm，a real—time hand tracking system for the stereo vision system of humanoid robot MIR一1 was presented． For each video frame，the raw image is converted to a 2一D skin color probability distribu—tion image via the skin c010r histogr。．!n back—projection． Camshift algorithm is then applied to find the mode(peak)of the probability distribution，which estimates the center and size of tracking window． To deal with the hand occlusion by other objects and choose the correct skin region when multiple image re—gions are skin colored，a two—dimensional Kalman filter(stochastic estimator)is used to track the hand re—gion centroid． The experimental results show that this algorithm is robust and efficient．
Key words：hand tracking；Camshift；Kalman f订ter；color probability distribution
页:
[1]