基于Mean Shift算法和Hough变换的疲劳驾驶检测方法  

A Method of Detecting Fatigue Driving Based on Mean Shift Algorithm and Hough Transformation Algorithm

作　　者：华国栋 徐志超[2] 叶鹏飞 

机构地区：[1]南京栎树交通互联科技有限公司,江苏南京211100 [2]常州工学院电气与光电工程学院,江苏常州213032

出　　处：《常州工学院学报》2017年第4期33-37,共5页Journal of Changzhou Institute of Technology

基　　金：国家科技支撑计划(2013BAH13F00);江苏省重点研发计划(产业前瞻与共性关键技术)重点项目(BE2017004-3);江苏省大学生创新创业训练计划项目(201611055056H)

摘　　要：依据PERCLOS基本原理,运用图像处理方法分析测量驾驶员疲劳状态的基本过程,提出一种采用Mean Shift算法和Hough变换的疲劳驾驶检测方法,给出了详细的实现步骤。实验表明,Mean Shift算法能对驾驶员眼部进行有效定位,Hough变换算法能正确判断眼睛闭合状态,所用方法对疲劳驾驶具有较好的检测效果。Based on the principleof PERCLOS,with an application of digital image processing technologies in analyzing the procedure of fatigue detection,a method of detecting fatigue driving has been proposed by combining Mean Shift algorithm and Hough transformation.Implementing steps are presented.The experiment results prove that eyes location can be realized by Mean Shift algorithm,and Hough transformation algorithm can determine the closed-open status of eyes.The method has a good effect in detecting fatigue in drivers.

关 键 词：PERCLOS MEANSHIFT 疲劳检测 HOUGH变换 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]