用红外图像实时跟踪和监测眼睛的方法

被引:11
作者
周玉彬
俞梦孙
机构
[1] 第四军医大学生物医学工程系
[2] 航空医学研究所 西安
[3] 北京
关键词
眼睛跟踪; 眼睛监测; 红外线;
D O I
暂无
中图分类号
R310 [];
学科分类号
1001 ;
摘要
眼睑和眼球状态是判断司机是否出现瞌睡和困倦的主要指标 ,其核心是实时地跟踪和检测眼睑的活动状态和眼球的注视位置 ,从而辨识出困倦程度 ,以便报警或提醒司机。当前较为理想的非接触式的检测途径是用摄像技术获取眼部的动态图像 ,并利用实时图像处理技术 ,跟踪和分析眼睑状态和眼球的注视位置 ,目前市场上尚无很成功的产品。本研究旨在探索利用摄像技术实时动态地提取眼睑和眼球状态的新方法。在 85 0nm的红外线照射下 ,利用瞳孔和虹膜对红外光吸收率的显著差别 ,引导图像处理区域改变 ,实现司机在一定范围内活动时跟踪眼睛 ,测量眼睑和眼球状态的分析技术。该方法为检测司机在驾驶中是否困倦提供了关键的技术 ,对驾驶员实际录像测量显示该方法是有效的。对汽车司机进行了 3次共 7 5h的实际驾车测验 ,跟踪准确率在97 5 % ,判断测量准确度为 97%。
引用
收藏
页码:104 / 108+103 +103
页数:6
相关论文
共 11 条
[1]  
Technologies for the monitoring and prevention of driver fatigue. Heitmann A, Guttkuhn R, Aguirre A, et al. Proceedings of the First International Driving Symposium on Human Factor in Driver Assessment,Training and Vehicle Design .
[2]  
Perclos:avalidpsychophysiologicalmeasureofalertnessasassessedbypsychomotorvigilance. DingesDF,GraceR. FederalHighwayAd ministration,OfficeofMotorCarriers,ReportNo.FHWA MCRT98006 . 1998
[3]  
Monitoringhead eyemotionfordriveralternesswithonecamera. SmithP,ShahM,LoboNV. ProceedingsoftheInternationalConferenceonPatternRecognition(ICPR’00) . 2000
[4]  
A drowsy driver detection system for heavy vehicles. Grace R, Byrne V E, Legrand J M, et al. Proceedings of the Digital Avionics Systems Conference (DASC), Belleview, Washington . 1998
[5]  
Development of driver alertness detection system using overhead capacitive sensor array. Philip K, Roger J, John D. . 1998
[6]  
Detecting and tracking eyes by using their physiological properties, dynamics, and appearance. Haro A, Flickner M, Essa I. . 2000
[7]  
3D Face pose estimation and tracking from a monocular camera. Ji Q, Hu R. Image and Vision Computing . 2002
[8]  
Effectivenessofpupilareadetectiontech niqueusingtwolightsourcesandimagedifferencemethod. EbisawaY,SatohS. Proceedingsofthe15thAnnualInt.Conf.oftheIEEEEnginMedicineandBiologySociety . 1993
[9]  
Unconstrained pupil detection technique using two lightsources and the image difference method. Ebisawa Y. Visualization and Intelli-gent Design in Engineering and Architecture . 1995
[10]  
Pupil detection and tracking using multiple. Morimoto C H, Koons D, Amir A, et al. . 1998