基于人眼状态识别的驾驶员疲劳检测方法研究  

Driver Fatigue Detection Based on Eye's State Recognition

作　　者：黄仁[1] 田源[1] 杨吉云[1] 孔润[1] 

机构地区：[1]重庆大学计算机学院,重庆400044

出　　处：《世界科技研究与发展》2011年第1期125-127,共3页World Sci-Tech R&D

基　　金：重庆市自然科学基金资助项目(2009BA2024)

摘　　要：提出一种基于眼睛状态识别的驾驶员疲劳状态检测的方法。首先通过建立肤色模型分割出人脸区域,再利用搜索连通域的方法实现眼睛区域的定位;然后计算出眼睛区域的垂直积分投影曲线,提取并比较曲线的特征参数,从而识别眼睛的状态;最后以眼睛闭合时间比率为测量标准实现对驾驶员疲劳程度的检测。实验表明,该方法简单、有效、实时性较好,疲劳检测的正确率较高。A detection method of driver fatigue state is proposed based on eye＇s state recognition. First,the face is divided through the establishment of skin model. The eye area is located by searching connected domains. Then vertical integral projection curve is measured to calculate the eye area. The curve＇s characteristic parameters are extracted and compared in order to identify the eye＇s state. Finally, driver fatigue is detected with the measurement standard of ratio of eyes closed time. The experiments show that this method is simple, effective and has good real - time performance. It is proved that the precision rate is high.

关 键 词：肤色模型 积分投影 眼睛状态识别 疲劳检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]