Fatigue features based on eye tracking for driver inattention system
Fatigue features based on eye tracking for driver inattention system
复制标题
基于眼动追踪的驾驶员注意力不集中系统的疲劳特征
DOI：
10.1109/tsp.2011.6043660
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
K. Horak
中科院分区：
文献类型：
--
作者：
K. Horak
本文研究了基于摄像机的驾驶员警觉性监控视觉系统的图像分割方法和疲劳特征的确定。一般来说，视觉监控系统必须分析一组计算出的疲劳特征，并识别驾驶员的注意力不集中或困倦。本文主要关注用于可靠眼动跟踪的分割方法，因为眼睛特征无疑是确定驾驶员疲劳的最重要特征。本文介绍了简单的颜色分割和霍夫变换等基本分割方法。然后介绍了一种更复杂的类哈尔特征方法和对称性检测方法。最后，列举并描述了几种主要的疲劳特征。所有的分割方法都在实验室和真实图像上进行了测试。
This paper deals with segmentation methods and fatigue features determination for a camera-based visual systems monitoring driver vigilance. Generally visual monitoring systems have to analyse a set of computed fatigue features and recognize driver inattention or sleepiness. The paper is focused mostly on the segmentation methods used for reliable eyes tracking because of eyes features are certainly the most significant features for determining of a driver fatigue. Fundamentals segmentation methods as a simple colour segmentation and Hough transform are introduced in the paper. After that a more complex Haar-like features approach and symmetries detection approach are shortly introduced. Finally, several of the leading fatigue features are listed and described. All the presented segmentation methods were tested on both laboratory and real images.