吉林大学学报(信息科学版)

• 论文 • 上一篇    下一篇

驾驶员疲劳状态检测方法研究

张雯頩1 ,康 冰2   

  1. 1. 吉林省科学技术协会学会服务中心,长春 130022; 2. 吉林大学 通信工程学院,长春 130022
  • 收稿日期:2017-12-07 出版日期:2018-05-24 发布日期:2018-07-25
  • 作者简介:张雯頩( 1983— ) ,女,吉林公主岭人,吉林省科学技术协会学会服务中心工程师,主要从事计算机管理与应用研究,( Tel) 86-13039308480( E-mail) 81604589@ qq. com; 康冰( 1978— ) ,男,吉林省吉林市人,吉林大学高级工程师,主要从事最优控制及模式识别研究,( Tel) 86-13514410736( E-mail) kangbing@ jlu. edu. cn。
  • 基金资助:
     国家自然科学基金资助项目( 61503151) ; 吉林省省级产业创新专项基金资助项目( 2017C032-4)

Research on Driver Fatigue State Testing Methods

ZHANG Wenping1 ,KANG Bing2   

  1. 1. Learning Service Center,Science and Technology Association of Jilin Province,Changchun 130022,China;2. College of Communication Engineering,Jilin University,Changchun 130022,China
  • Received:2017-12-07 Online:2018-05-24 Published:2018-07-25

摘要: 为解决由于疲劳驾驶导致交通事故的问题,采用视频图像分析技术处理疲劳的相关特征,运用基于训练的 Adaboost 人脸检测算法精确定位司机脸部和眼睛区域,实时采集眼睛二值化区域面积,采用阈值比较法进行眨眼判断,并提取眼皮疲劳参数 AECS( Average Eyelid Closing Speed) 和 PERCLOS( Percent Eyelid Closure over the Pupil Time) ,进行综合疲劳状态分析,最终确定是否疲劳驾驶。实验结果显示,人脸和人眼检测的精度都有较大程度提高,设计的软件可实时监测驾驶员疲劳状态,有效防止疲劳驾驶。

关键词: 图像预处理, Adaboost 算法, 人脸检测, 疲劳驾驶

Abstract: To solve the problem of traffic accidents due to fatigue driving,we use video image processing related characteristics of fatigue analysis technology,use Adaboost face detection algorithm based on training precise positioning driver face and eye area,real-time acquisition eyes area binarization,a judging threshold value comparison method is adopted to improve the blink of an eye,and extract the AECS( Average Eyelid Closing Speed) eyelid fatigue parameters, the PERCLOS ( Percent of Eyelid Closure over the Pupil Time ) , a comprehensive state of fatigue analysis,ultimately determine whether fatigue driving. The experiment results show that the precision of the face and eye detection has greatly improved,the designed software can detect the
driver's fatigue state by real time and avoid fatigue driving.

Key words: Adaboost, face detection, image preprocessing, fatigue driving

中图分类号: 

  • TP391. 41