改进CamShift算法的眼动跟踪方法

Eye tracking method on improved CamShift algorithm

针对CamShift跟踪算法仅采用颜色特征,在存在颜色相近干扰目标、头部快速运动或者虹膜发生形变等情况易发生眼动跟踪不准确或失败等问题,提出一种基于改进CamShift算法的眼动跟踪方法。在CamShift算法中,计算边缘直方图分布,在颜色特征基础上融合边缘特征,同时通过分析饱和度分量抑制噪声影响,并利用基于边缘特征的自定义方法判断虹膜是否发生形变。当发生形变时,自动更新模板,根据历史运动轨迹预测虹膜中心。实验表明,该方法能有效改善眼动跟踪性能,且定位虹膜精确、错误率低、速度快,预测虹膜中心与实际虹膜中心相差极小,达到了准确性、鲁棒性和实时性的要求。

Since the CamShift algorithm only adopted color feature, which would result in the inaccuracy or failure of eye tracking in the Situation of similar color objects, rapid movement of the head or deformation of the iris, this paper proposed an eye tracking method based on improved CamShift algorithm. In the CamShift algorithm, it calculated edge histogram distribution and carried out a fusion of edge feature and color feature, at the same time, by analyzing the saturation component, suppressed noise, and used a custom criteria based on edge feature to determine the iris deformation. When it happened, automatically updated the template and predicted the center of iris according to its historical trajectory. The experimental results show the proposed algorithm can improve the performance of eye tracking effectively and accurately track iris with low error rate, high speed and tiny gap between the pre- dicted center of iris and its actual center. It meets the requirements of accuracy, robustness and real-time.