摘要
针对在实际环境中疲劳驾驶检测系统对于实时性和准确性的要求,设计了一个基于DSP的疲劳驾驶视觉检测与预警系统.系统首先利用CMOS摄像头MT9V034采集驾驶员的图像,然后通过AdaBoost算法在图像中检测出人脸,并在检测到的人脸区域对驾驶员的人眼进行定位.最后,在对人眼图像特征分析的基础上使用统计学的方法实现眼睛闭合度的计算,从而检测出驾驶员的疲劳程度并进行预警.实验表明,与传统的疲劳检测方法相比,本系统在一定程度上提高了疲劳检测的实时性和准确性,适合应用在实际行驶过程中.
Aiming at the real-time and accuracy requirements of fatigue driving detection system in reality,a DSP-based visual inspection and early warning system for fatigue driving was designed.The system primarily uses the CMOS camera MT9V034 to capture the driver’s image,and the AdaBoost algorithm to detect the human face in the image,and locates the driver’s eyes in the detected face area.Based on the analysis of the features of the human eye image,a statistical method is used to calculate the degree of eye closure,thereby detecting the driver’s fatigue level and providing an early warning.Experiments show that compared with traditional fatigue detection methods,this system improves the real-time performance and accuracy of fatigue detection,and is suitable for application in actual driving process.
作者
王欣
吴键
孙涵
周建荣
WANG Xin;WU Jian;SUN Han;ZHOU Jianrong(School of Mechanical and Engineering,NJUST,Nanjing 210094,China;Yangzhou Zilu Information Technology Co.,Ltd.,Yangzhou 225000,China)
出处
《测试技术学报》
2020年第6期506-513,共8页
Journal of Test and Measurement Technology
关键词
疲劳驾驶
机器学习
人脸检测
人眼定位
嵌入式系统
fatigue driving
machine learning
face detection
eye positioning
embedded systems
作者简介
王欣(1994-),男,硕士生,主要从事机器视觉与图像处理、嵌入式系统等研究.
