基于视觉的疲劳驾驶监测关键技术研究
|
摘要
由于疲劳驾驶导致的交通事故数量不断上升，疲劳驾驶已成为一个严重的社会问题。因此，研制疲劳驾驶监测预警系统，对于避免交通事故，提高交通系统的安全性有着重要意义。本文重点研究了基于视觉的疲劳驾驶监测系统中的一些关键技术，并开发了一套疲劳监测软件系统。
本文首先对传统的眼睛模型进行了改进，采用上眼睑高度来判断眼睛状态。
其次，研究了眼睛的定位和眼睛特征的提取。在眼睛定位方面，采用了两种实时性较高的算法，形态学方法和积分投影法，对这两种算法在准确率和检测耗时方面进行比较后，确定采用积分投影法。在眼睛特征的提取中，研究了活动轮廓模型(Snake)提取上眼睑，在对上眼睑特征分析的基础上对传统的活动轮廓模型进行了改进，提出了分阶段的活动轮廓模型生长方法，该方法有效地利用了全局信息，减少了检测时间。实验证明该方法在提取上眼睑时能够适应各种眼睛状态，具有较高的正确率。
然后，提取了三个疲劳参数，PERCLOS，Eye Closure Time(眼睛闭合时间)和Eye Blink Frequency(眨眼频率)。对于三个疲劳参数的融合，研究采用基于ID3算法的决策树，初步实现了三个疲劳参数的决策树生成和疲劳识别。
最后，基于上述研究实现了一套疲劳监测软件系统，通过实验证明了系统能够实时地监测驾驶者的眼睛状态并进行疲劳识别，在实验室环境下有较高的识别率。
The increasing number of traffic accidents due to a driver's fatigue has become aserious problem for the society. With the background, developing a system for monitoringthe driver's level of vigilance and alerting the driver when he fatigues is essential. In thisthesis, a software system is developed for detecting the driver's fatigue and some keytechnologies are presented.
First of all, a new eye model is proposed in our system. According to the model, theheight of upper eyelid is detected to show the eyes' status.
Secondly, the eyes' locations are found and some features of eyes are detected. Inorder to get the locations of the eyes, two methods are implemented, Morphologicaltransformation algorithm and Grayness integral projection algorithm. The precision andtime cost for the two methods are compared and the latter is better for our system. Activecontour model, which is also named Snake algorithm, is introduced to detect the uppereyelid. Because of the special grads of the eye, an ameliorated snake algorithm is proposed.The new algorithm consists two phases. The first phase utilizes the overall information ofthe edge of the eye to be detected. The second phase utilizes the grads of the eye image.The experimental results show that the new method could detect the upper eyelideffectively.
Finally, the three fatigue parameters are picked up, which are PERCLOS, Eye closuretime and Eye blink frequency. ID3 algorithm is introduced to develop a decision tree whichcan use the three parameters to recognize the driver's fatigue.
A software system based on the above technologies is implemented and the system candetect the driver's eyes and recognize his fatigue in real time. From the running effect ofthis system, the efficiency of this system is very high in the laboratory environment.
本文首先对传统的眼睛模型进行了改进，采用上眼睑高度来判断眼睛状态。
其次，研究了眼睛的定位和眼睛特征的提取。在眼睛定位方面，采用了两种实时性较高的算法，形态学方法和积分投影法，对这两种算法在准确率和检测耗时方面进行比较后，确定采用积分投影法。在眼睛特征的提取中，研究了活动轮廓模型(Snake)提取上眼睑，在对上眼睑特征分析的基础上对传统的活动轮廓模型进行了改进，提出了分阶段的活动轮廓模型生长方法，该方法有效地利用了全局信息，减少了检测时间。实验证明该方法在提取上眼睑时能够适应各种眼睛状态，具有较高的正确率。
然后，提取了三个疲劳参数，PERCLOS，Eye Closure Time(眼睛闭合时间)和Eye Blink Frequency(眨眼频率)。对于三个疲劳参数的融合，研究采用基于ID3算法的决策树，初步实现了三个疲劳参数的决策树生成和疲劳识别。
最后，基于上述研究实现了一套疲劳监测软件系统，通过实验证明了系统能够实时地监测驾驶者的眼睛状态并进行疲劳识别，在实验室环境下有较高的识别率。
The increasing number of traffic accidents due to a driver's fatigue has become aserious problem for the society. With the background, developing a system for monitoringthe driver's level of vigilance and alerting the driver when he fatigues is essential. In thisthesis, a software system is developed for detecting the driver's fatigue and some keytechnologies are presented.
First of all, a new eye model is proposed in our system. According to the model, theheight of upper eyelid is detected to show the eyes' status.
Secondly, the eyes' locations are found and some features of eyes are detected. Inorder to get the locations of the eyes, two methods are implemented, Morphologicaltransformation algorithm and Grayness integral projection algorithm. The precision andtime cost for the two methods are compared and the latter is better for our system. Activecontour model, which is also named Snake algorithm, is introduced to detect the uppereyelid. Because of the special grads of the eye, an ameliorated snake algorithm is proposed.The new algorithm consists two phases. The first phase utilizes the overall information ofthe edge of the eye to be detected. The second phase utilizes the grads of the eye image.The experimental results show that the new method could detect the upper eyelideffectively.
Finally, the three fatigue parameters are picked up, which are PERCLOS, Eye closuretime and Eye blink frequency. ID3 algorithm is introduced to develop a decision tree whichcan use the three parameters to recognize the driver's fatigue.
A software system based on the above technologies is implemented and the system candetect the driver's eyes and recognize his fatigue in real time. From the running effect ofthis system, the efficiency of this system is very high in the laboratory environment.
引文
1．智能运输系统和GPS车辆导航系统[Online]．http://publish.it168.com/2005/0730/20050730017701.shtml, 2005-7-30.
2．郭克友．驾驶员疲劳状态视觉监测技术研究[D]．吉林：吉林大学，2003．
3．现代智能车辆的研究与发展史[Online]．http://www.pcauto.com.cn/teach/qczs/0409/125654.html, 2004-11-8.
4. Gerry E. Warning system for fatigued drivers nearing reality with new eye data[J]. Science Daily Magazine, 1999(7):25～30.
5. C.J. Reissman. The Alert Driver: A Trucker's Guide to Sleep, Fatigue, and Rest in our 24-Hour Society. American Trucking Associations, 2200 Mill Road, Alexandria, USA, 1996.
6. D. Royal. Volume I—Findings report; national survey on distracted and driving attitudes and behaviors. The Gallup Organization, Washington, D.C., Tech. Rep. DOT HS 809 566, Mar.2003.
7. Luis M. Bergasa. Real-Time System for Monitoring Driver Vigilance. IEEE Transaction on Intelligent Transportation Systems[J], 2006, 7(6): 63-77.
8．毛喆等．汽车驾驶员驾驶疲劳监测技术研究进展[J]．中国安全科学学报，2005，15(3)：108～112．
9．张灵聪，王正国，朱佩芳，尹志勇．汽车驾驶疲劳研究综述[J]．人类工效学，2003，9(1)：39～42．
10. Daimler Chrysler AG. The Electronic Drawbar. [Online].Available: http://www.daimlerchrysler.com, 2001, Jun.
11. Dinges David F.Ph.D., Grace Richard Ph.D. PERCLOS: A Valid Psycho physiological Measure of Alertness Assessed by Psychomotor Vigilance. Federal Highway Administration, Office of Motor Carriers, 1998.
12. Grace. R., Benjamin A.L. Application of a Heavy Vehicle Drowsy Driver Detection System [A]. In: Proceedings of the SAE International Truck & Bus Meeting and Exposition[C], November 15-17, 1999.
13. Y. Matsumoto and A. Zelinsky. An algorithm for real-time stereo vision implementation of head poses and gaze direction measurements [A]. In: Proc. IEEE 4th Int. Conf. Face and Gesture Recognition[C], Grenoble, France, Mar, 2000: 499-505.
14. S. Boverie, J. M. Leqellec, and A.Hirl. Intelligent systems for video monitoring of vehicle cockpit [A]. In: Proc. Int. Cong. And Expo. ITS—Advanced Controls and Vehicle Navigation Systems, Detroit[C], MI, Feb, 1998:1-5.
15. W. Shih and J. Liu. A calibration-free gaze tracking technique[A]. In: Proc. 15th Conf. Patterns Recognition, Barcelona[C], Spain, 2000, vol(4): 201-204.
16. Q, Ji and X. Yang. Real-time eye, gaze and face pose tracking for monitoring driver vigilance[J]. Real-Time Imaging, 8(5), 2002: 357-377.
17. A. Polychronopoulos, A. Amditis, and E. Bekiaris. Information data flow in awake multi-sensor driver monitoring system[A]. In: Proc. IEEE Intelligent vehicles Symp[C], Parma, Italy, Jun. 2004: 902-906.
18．王荣本，郭克友等．一种基于Gabor小波的驾驶员眼部状态识别方法的研究[J]．中国图像图形学报，2003，8A(9)：1043-1047．
19．李锋．机器视觉应用技术研究[D]．浙江：浙江大学，2003．
20．陈艳琴．关于司机疲劳监测的人眼监测与跟踪研究[D]．长沙：中南大学，2004．
21．林维训，潘钢，吴朝晖，潘云鹤．脸部特征定位方法fJ]．中国图像图形学报，2003，8A(8)：849-859．
22. Zhang L M, Lenders P. Knowledge-based eye detection for human face recognition[A]. In: Proceedings of Knowledge-Based Intelligent Engineering Systems and Allied Technologies[C], Brighton, UK, 2000, 1: 117-120.
23．山世光，高文，陈熙霖。基于纹理分布和变形模板的面部特征提取[J]．软件学报，2001，12(4)：570-577．
24．王磊，莫玉龙，戚飞虎．基于霍夫变换和眼睑弹性模板的眼睛特征提取[J]．红外与毫米波学报，1999，18(1)：53-60．
25．Milan Sonka，Vaclav Hlavac，Roger Boyle．图像处理，分析与机器视觉[M]．第二版．北京：人民邮电出版社．2003．
26. Yuille A. L., Cohen D. S., Halliman PW. Feature extraction from faces using deformable templates[A]. In: Proceedings of Computer Vision and Pattern Recognition[C], San Diego, CA, U SA, 1989: 104-109.
27. Cai J. Goshtasby. A Detecting face in color images, Image and Vision Computing[J], 1999, 18(1): 63-75.
28. Zhou jie, Lu Chun-Yu, Zhang Chang-Shui. Based on directional symmetry transform. Et al. Human face locations Acta Electronica Sinica, 1999 27(8): 12-15.
29．刘文予，潘峰。离散对称变换在人脸图像眼睛定位中的应用[J]．红外与毫米波学报，2001，20(5)：375-380。
30. Daugman. Biometric personal identification system based on iris analysis[P]. US. Pattent 5291560. 1994.
31. Freedman H., Davis L S. A corner finding algorithm for chain code curves. IEEE Trans.On Computers, 1997, 26: 297～303.
32. Wand H, Brany M. Real-time comer detection algorithm for motion estimation. Image and Vision Computer. 1995, (9): 695～703.
33．廖常俊等．一种改进的角点检测法．仪器仪表用户，2005，12(2)：81～83．
34．张坤华，王敬儒，张启衡．多特征复合的角点提取方法[J]．中国图像图形学报，2002，7(1)：319～324．
35．陈白帆 蔡自兴．基于尺度空间理论的Harris角点检测．中南大学学报，2005，36(5)：29-34．
36．陈国勇．汽车安全驾驶中的视线估计[D]．南京：南京理工大学，2006。
37. Luis Jordao, Matteo Perrone and Joao Paulo Costeira. Active Face and Feature Tracking. Proc. of IEEE Conference on Computer Vision and Pattern Recognition, 1999.
38. Rein-Lien Hsu. Face Detection in Color Images. IEEE Transaction on Pattern Analysis and Machine Intelligence. 2002, 24(5): 696-706.
39. Kanade T. Picture processing by computer complex and recognition of human faces[Ph.D. Thesis]. Kyoto: Kyoto University, 1973.
40. Feng GC, Yuen PC. Variance projection function and its application to eye detection for human face recognition. Pattern Recognition Letters, 1998, 19(9): 899-906.
41．耿新，周志华。基于混合投影函数的眼睛定位[J]．软件学报，2003，14(8)．
42. Kass M, WitKinA, Terzopoulous D. Snakes: active contour models. Proceedings of the 1st International Conference on Computer Vision, IEEE Computer Society Press, 1987, 259～268.
43．李培华，张田文．活动轮廓模型综述[J]．软件学报，2001，11(6)：751～757．
44. Williams D J, Shab M. A fast algorithm for active contours and curvature estimation. CVGIP: Image Understanding, 1992, 55(1): 14～26.
45. Jian-Gang Wang, Eric Sung, Ronda Venkateswarlu. Eye Gaze Estimation from a Single Image of One Eye. Proceeding of the Ninth IEEE International Conference on Computer Vision. 2003.
46．Tom M．Mitchell．Machine Learning[M]．机械工业出版社．2004．
47．刘小虎，李生．决策树的优化算法[J]．软件学报，1998，9(10)：797～800．
48．John Durkin，蔡竞峰，蔡自兴．决策树技术及其当前研究方向[J]．控制工程，2005，12(1)：15-21．
49．王曙燕．医学图像智能分类算法研究[D]．西安：西北大学博士论文，2006．
2．郭克友．驾驶员疲劳状态视觉监测技术研究[D]．吉林：吉林大学，2003．
3．现代智能车辆的研究与发展史[Online]．http://www.pcauto.com.cn/teach/qczs/0409/125654.html, 2004-11-8.
4. Gerry E. Warning system for fatigued drivers nearing reality with new eye data[J]. Science Daily Magazine, 1999(7):25～30.
5. C.J. Reissman. The Alert Driver: A Trucker's Guide to Sleep, Fatigue, and Rest in our 24-Hour Society. American Trucking Associations, 2200 Mill Road, Alexandria, USA, 1996.
6. D. Royal. Volume I—Findings report; national survey on distracted and driving attitudes and behaviors. The Gallup Organization, Washington, D.C., Tech. Rep. DOT HS 809 566, Mar.2003.
7. Luis M. Bergasa. Real-Time System for Monitoring Driver Vigilance. IEEE Transaction on Intelligent Transportation Systems[J], 2006, 7(6): 63-77.
8．毛喆等．汽车驾驶员驾驶疲劳监测技术研究进展[J]．中国安全科学学报，2005，15(3)：108～112．
9．张灵聪，王正国，朱佩芳，尹志勇．汽车驾驶疲劳研究综述[J]．人类工效学，2003，9(1)：39～42．
10. Daimler Chrysler AG. The Electronic Drawbar. [Online].Available: http://www.daimlerchrysler.com, 2001, Jun.
11. Dinges David F.Ph.D., Grace Richard Ph.D. PERCLOS: A Valid Psycho physiological Measure of Alertness Assessed by Psychomotor Vigilance. Federal Highway Administration, Office of Motor Carriers, 1998.
12. Grace. R., Benjamin A.L. Application of a Heavy Vehicle Drowsy Driver Detection System [A]. In: Proceedings of the SAE International Truck & Bus Meeting and Exposition[C], November 15-17, 1999.
13. Y. Matsumoto and A. Zelinsky. An algorithm for real-time stereo vision implementation of head poses and gaze direction measurements [A]. In: Proc. IEEE 4th Int. Conf. Face and Gesture Recognition[C], Grenoble, France, Mar, 2000: 499-505.
14. S. Boverie, J. M. Leqellec, and A.Hirl. Intelligent systems for video monitoring of vehicle cockpit [A]. In: Proc. Int. Cong. And Expo. ITS—Advanced Controls and Vehicle Navigation Systems, Detroit[C], MI, Feb, 1998:1-5.
15. W. Shih and J. Liu. A calibration-free gaze tracking technique[A]. In: Proc. 15th Conf. Patterns Recognition, Barcelona[C], Spain, 2000, vol(4): 201-204.
16. Q, Ji and X. Yang. Real-time eye, gaze and face pose tracking for monitoring driver vigilance[J]. Real-Time Imaging, 8(5), 2002: 357-377.
17. A. Polychronopoulos, A. Amditis, and E. Bekiaris. Information data flow in awake multi-sensor driver monitoring system[A]. In: Proc. IEEE Intelligent vehicles Symp[C], Parma, Italy, Jun. 2004: 902-906.
18．王荣本，郭克友等．一种基于Gabor小波的驾驶员眼部状态识别方法的研究[J]．中国图像图形学报，2003，8A(9)：1043-1047．
19．李锋．机器视觉应用技术研究[D]．浙江：浙江大学，2003．
20．陈艳琴．关于司机疲劳监测的人眼监测与跟踪研究[D]．长沙：中南大学，2004．
21．林维训，潘钢，吴朝晖，潘云鹤．脸部特征定位方法fJ]．中国图像图形学报，2003，8A(8)：849-859．
22. Zhang L M, Lenders P. Knowledge-based eye detection for human face recognition[A]. In: Proceedings of Knowledge-Based Intelligent Engineering Systems and Allied Technologies[C], Brighton, UK, 2000, 1: 117-120.
23．山世光，高文，陈熙霖。基于纹理分布和变形模板的面部特征提取[J]．软件学报，2001，12(4)：570-577．
24．王磊，莫玉龙，戚飞虎．基于霍夫变换和眼睑弹性模板的眼睛特征提取[J]．红外与毫米波学报，1999，18(1)：53-60．
25．Milan Sonka，Vaclav Hlavac，Roger Boyle．图像处理，分析与机器视觉[M]．第二版．北京：人民邮电出版社．2003．
26. Yuille A. L., Cohen D. S., Halliman PW. Feature extraction from faces using deformable templates[A]. In: Proceedings of Computer Vision and Pattern Recognition[C], San Diego, CA, U SA, 1989: 104-109.
27. Cai J. Goshtasby. A Detecting face in color images, Image and Vision Computing[J], 1999, 18(1): 63-75.
28. Zhou jie, Lu Chun-Yu, Zhang Chang-Shui. Based on directional symmetry transform. Et al. Human face locations Acta Electronica Sinica, 1999 27(8): 12-15.
29．刘文予，潘峰。离散对称变换在人脸图像眼睛定位中的应用[J]．红外与毫米波学报，2001，20(5)：375-380。
30. Daugman. Biometric personal identification system based on iris analysis[P]. US. Pattent 5291560. 1994.
31. Freedman H., Davis L S. A corner finding algorithm for chain code curves. IEEE Trans.On Computers, 1997, 26: 297～303.
32. Wand H, Brany M. Real-time comer detection algorithm for motion estimation. Image and Vision Computer. 1995, (9): 695～703.
33．廖常俊等．一种改进的角点检测法．仪器仪表用户，2005，12(2)：81～83．
34．张坤华，王敬儒，张启衡．多特征复合的角点提取方法[J]．中国图像图形学报，2002，7(1)：319～324．
35．陈白帆 蔡自兴．基于尺度空间理论的Harris角点检测．中南大学学报，2005，36(5)：29-34．
36．陈国勇．汽车安全驾驶中的视线估计[D]．南京：南京理工大学，2006。
37. Luis Jordao, Matteo Perrone and Joao Paulo Costeira. Active Face and Feature Tracking. Proc. of IEEE Conference on Computer Vision and Pattern Recognition, 1999.
38. Rein-Lien Hsu. Face Detection in Color Images. IEEE Transaction on Pattern Analysis and Machine Intelligence. 2002, 24(5): 696-706.
39. Kanade T. Picture processing by computer complex and recognition of human faces[Ph.D. Thesis]. Kyoto: Kyoto University, 1973.
40. Feng GC, Yuen PC. Variance projection function and its application to eye detection for human face recognition. Pattern Recognition Letters, 1998, 19(9): 899-906.
41．耿新，周志华。基于混合投影函数的眼睛定位[J]．软件学报，2003，14(8)．
42. Kass M, WitKinA, Terzopoulous D. Snakes: active contour models. Proceedings of the 1st International Conference on Computer Vision, IEEE Computer Society Press, 1987, 259～268.
43．李培华，张田文．活动轮廓模型综述[J]．软件学报，2001，11(6)：751～757．
44. Williams D J, Shab M. A fast algorithm for active contours and curvature estimation. CVGIP: Image Understanding, 1992, 55(1): 14～26.
45. Jian-Gang Wang, Eric Sung, Ronda Venkateswarlu. Eye Gaze Estimation from a Single Image of One Eye. Proceeding of the Ninth IEEE International Conference on Computer Vision. 2003.
46．Tom M．Mitchell．Machine Learning[M]．机械工业出版社．2004．
47．刘小虎，李生．决策树的优化算法[J]．软件学报，1998，9(10)：797～800．
48．John Durkin，蔡竞峰，蔡自兴．决策树技术及其当前研究方向[J]．控制工程，2005，12(1)：15-21．
49．王曙燕．医学图像智能分类算法研究[D]．西安：西北大学博士论文，2006．