• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2006, Vol. 28 ›› Issue (11): 27-29.

• 论文 • 上一篇    下一篇

基于规则与卡尔曼滤波的人眼跟踪

董文会[1] 吴晓娟[1] 曲培树[2]   

  • 出版日期:2006-11-01 发布日期:2010-05-20

  • Online:2006-11-01 Published:2010-05-20

摘要:

本文提出一种基于YCbCr空间、规则和卡尔曼滤波的人眼跟踪方法。该方法首先利用肤色在YCbCr空间的特性建立肤色模型分割出肤色,然后通过设定的规则得到人眼区域,最后利用卡尔曼滤波的方法实时跟踪人眼。实验表明,该方法能达到实时效果且跟踪正确率较高。

关键词: YCbCr空间 肤色分割 人眼定位 卡尔曼滤波

Abstract:

An eye tracking method based on the YCbCr color space, rules and the Kalman filtering is proposed in this paper. First, all skin color regions can be  got from a skin model in the YCbCr color space, then the eye regions can be found according to some rules. Finally, the eye position is tracked by the K  alman filtering. Experimental results show that the method can track the eye in a real-time way and has very high correctness rate.

Key words: YCbCr color space;skin color segmentatiomeye locatiomKalman filtering