基于动态匹配模型的驾驶人嘴部行为识别与分级预警

被引:4
作者
付锐 [1 ]
程文冬 [2 ,3 ]
张名芳 [2 ]
袁伟 [2 ]
刘卓凡 [2 ]
郭艳君 [2 ]
机构
[1] 长安大学汽车运输保障技术交通行业重点实验室
[2] 长安大学汽车学院
[3] 西安工业大学机电工程学院
关键词
嘴部行为; 高斯混合模型; 能量角点; 动态匹配模型; 分级预警;
D O I
10.19562/j.chinasae.qcgc.2015.09.019
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
鉴于嘴部行为能够反映出驾驶人不同的精神状态,本文中提出了一种基于机器视觉的嘴部行为识别与精神状态预警方法。首先采用改进的高斯混合模型进行光照自适应的唇色分割。然后在嘴唇区域内建立了基于灰度能量角点的嘴唇动态匹配模型,用于描述嘴部特征并提取动作参数。最终建立了嘴部行为的判别模式,实现了驾驶人精神状态的分级预警。实验结果表明,动态匹配模型具有较高的嘴部特征匹配精度和动作适应能力。算法能够有效检测"闭嘴"、"说话"和"打哈欠"行为,分心预警和疲劳预警的监测精度分别为85.8%和92.9%。
引用
收藏
页码:1095 / 1102
页数:8
相关论文
共 17 条
[1]  
Efficient facial landmark localization using spatial–context AdaBoost algorithm[J] . Sungjin Hong,Sarang Khim,Phill Kyu Rhee. &nbspJournal of Visual Communication and Image Representation . 2014
[2]   Region duplication detection based on Harris corner points and step sector statistics [J].
Chen, Likai ;
Lu, Wei ;
Ni, Jiangqun ;
Sun, Wei ;
Huang, Jiwu .
JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2013, 24 (03) :244-254
[3]   Is a hands-free phone safer than a handheld phone? [J].
Ishigami, Yoko ;
Klein, Raymond M. .
JOURNAL OF SAFETY RESEARCH, 2009, 40 (02) :157-164
[4]   A comparative study of facial appearance modeling methods for active appearance models [J].
Lee, Sung Joo ;
Park, Kang Ryoung ;
Kim, Jaihie .
PATTERN RECOGNITION LETTERS, 2009, 30 (14) :1335-1346
[5]  
A survey of skin-color modeling and detection methods[J] . P. Kakumanu,S. Makrogiannis,N. Bourbakis. &nbspPattern Recognition . 2006 (3)
[6]  
Fully automated real time fatigue detection of drivers through Fuzzy Expert Systems[J] . Tayyaba Azim,M. Arfan Jaffar,Anwar M. Mirza. &nbspApplied Soft Computing Journal . 2014
[7]   A variational framework for Retinex [J].
Kimmel, R ;
Elad, M ;
Shaked, D ;
Keshet, R ;
Sobel, I .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2003, 52 (01) :7-23
[8]   Adaptive skin-color filter [J].
Cho, KM ;
Jang, JH ;
Hong, KS .
PATTERN RECOGNITION, 2001, 34 (05) :1067-1073
[9]  
Face detection based on skin color likelihood. Yusseok Ban,Sang-Ki Kim,Sooyeon Kim, et al. Pattern Recognition . 2014
[10]   汽车技术的发展动向及我国的对策 [J].
李克强 .
汽车工程, 2009, 31 (11) :1005-1016