In order to improve the defects of current eye state detection when used for fatigue monitoring, this paper proposes a new vision sensor-based comparison of eye gray area and eyelid curvature. Under the premise that the eyes area is found, the eye state is determined by extracting the features of the gray area and eyelid curvature characteristic features. Eye state detection in such complex environments as different backgrounds, changing illumi- nations, eye tilting and rotation, and wearing glasses shows that the algorithm has faster processing speed and ro- bustness.%针对眼睛状态检测在实际疲劳监控系统应用中的缺陷,提出了一种基于区域灰度特征比较和眼睑曲率的人眼状态识别算法。在确定眼睛区域的前提下,提取眼睛区域的灰度特征和眼睑曲率特征,确定眼睛状态。通过对不同背景、光照变化、眼睛倾斜和旋转,以及戴眼镜等多种复杂环境下进行眼睛状态检测,实验结果表明,文中算法具有处理速度快、鲁棒性高等特点。
展开▼