首页 | 本学科首页   官方微博 | 高级检索  
     

基于图像处理的驾驶员疲劳检测方法
引用本文:董文会,吴晓娟,徐祗军.基于图像处理的驾驶员疲劳检测方法[J].计算机应用与软件,2006,23(12):70-71,77.
作者姓名:董文会  吴晓娟  徐祗军
作者单位:1. 德州学院物理系,山东,德州,253023;山东大学信息科学与工程学院,山东,济南,250100
2. 山东大学信息科学与工程学院,山东,济南,250100
摘    要:提出一种基于图像处理的实时检测驾驶疲劳的方法,利用肤色信息、投影和连通域相结合的方法检测眼睛,通过动态模板匹配的方法对眼睛进行跟踪;最后利用上下眼睑距离来判断是否疲劳。实验表明,该方法可使眼睛定位和追踪的平均正确率达98%,疲劳检测的正确率为100%,准确率平均为90%。

关 键 词:疲劳检测  人脸定位  眼睛追踪  动态模板匹配
收稿时间:2005-03-08
修稿时间:2005-03-08

DRIVER FATIGUE DETECTION BASED ON IMAGE PROCESSING
Dong Wenhui,Wu Xiaojuan,Xu Zhijun.DRIVER FATIGUE DETECTION BASED ON IMAGE PROCESSING[J].Computer Applications and Software,2006,23(12):70-71,77.
Authors:Dong Wenhui  Wu Xiaojuan  Xu Zhijun
Affiliation:1.Department of Physics, Dezhou University, Dezhou Shandong 253023, China;2.School of Information Science and Engineering,Shandong University, Jinan Shangdong 250100, China
Abstract:A driver fatigue detection method based on image processing is proposed. The eyes are detected using the characteristics of skin colors, projections and finding connected components and are used as the dynamic templates for eye tracking in the next frame. Finally, driver fatigue is detected by the distance change of the eyelid. The average correct rate for eye location and tracking can reach 98% ,the correct rate for fatigue detection is 100% ,the average precision rate is 90%.
Keywords:Fatigue detection Face location Eye tracking Dynamic template matching
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号