基于行为特征的疲劳驾驶检测技术研究  

Research on Driver Fatigue Detection Based on Behavior Characteristics

作　　者：陈诗礼[1,2] 雷霖[2] 赵永鑫[2] 

机构地区：[1]西华大学数学与计算机学院,四川成都610039 [2]成都大学信息科学与技术学院,四川成都610106

出　　处：《成都大学学报（自然科学版）》2014年第1期37-40,共4页Journal of Chengdu University（Natural Science Edition）

基　　金：四川省科技厅基础应用研究(2013JY0117)资助项目

摘　　要：介绍了几类目前常用的疲劳检测技术的优缺点,提出了一种改进的疲劳驾驶检测方法:先通过2次图像投影和形态学方法实现眼睛精确定位;再根据眼睛睁闭时,其眼睛宽高比的差异,提出一种眼睛状态的识别方法;根据PERCLOS方法的判断是否疲劳.算法能够有效减少计算量提高运算速度,并在实验室内取得了较高的精确度.This paper discusses the advantages and disadvantages of some common methods about driver fa- tigue detection, and then proposes an improved method. Firstly, the method uses twice image projection and morphologic method to achieve precise eye location. Then, according to the difference of the width-height ratio of the eyes while opening and closing eyes, one new way for eye state identification is proposed. Final- ly, according to PERCLOS method, we can determine whether fatigue shows on driver. The algorithm can ef- fectively reduce the amount of calculation, and improve computing speed, and achieve high accuracy in the laboratory.

关 键 词：疲劳驾驶检测 图像投影 眼睛定位 眼睛状态识别 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]