Article Details

Abstract

To address the frequent traffic accidents caused by fatigue driving, this paper designs a fatigue driving detection system based on multi-feature fusion. The system utilizes Dlib and OpenCV to extract key features of drivers, including eye-opening degree, mouth action units, and head posture angles. These features are integrated with machine learning algorithms to construct a classification model for identifying fatigue states. Experimental results demonstrate that the method achieves an accuracy rate of 97.6% on a self-built dataset, significantly enhancing the robustness of fatigue detection. The research validates the application value of multi-feature fusion strategies in driving behavior monitoring and provides technical support for mitigating traffic accident risks.

Keywords

fatigue driving detection; facial feature; feature extraction; multi-feature fusion

References

1. [1] Chowdhury A，Shankaran R，Kavakli M，et al. Sensor Applications and Physiological Features in Drivers' Drowsiness Detection: A Review [J].IEEE Sensors Journal，2018，18（8）：3055-3067.
2. [2] Li Ying, Lai Jianhuang, Ruan Bangzhi Combination method for eye state detection [J]. Chinese Journal of Image and Graphics, 2003 (10): 1140-1145
3. [3] Abdel-Mottaleb M，Elgammal A. Face Detection in Complex Environments from Color Images [C]//IEEE International Conference on Image Processing，1999，3：622-626.
4. [4] Zhang F，Su J J，Geng L，et al. Driver Fatigue Detection Based on Eye State Recognition [C]//2017 International Conference on Machine Vision and Information Technology (CMVIT)，2017：105-110.
5. [5] Su Jinjin, Huo Chunbao, Wang Te Driver fatigue detection system based on facial features [J]. Information Technology and Informatization, 2021 (9): 114-116
6. [6] Ma Xueting, Fei Shumin Research on Fatigue Driving State Detection Based on Facial Features and Deep Learning [J]. Electronic Testing, 2021 (11): 33-36
7. [7] Hu Xizhi, Huang Bingyu Fatigue driving detection method based on facial feature analysis [J]. Science, Technology and Engineering, 2021, 21 (4): 1629-1636
8. [8] Zhang Xuxin, Wang Xuesong, Ma Yong, etc International research progress on driving behavior and driving risk [J]. Chinese Journal of Highways, 2020, 33 (6): 1-17
9. [9] Sun Peng, Sun Chuancong, Xu Yaoyao, wait Fatigue driving monitoring and alarm system based on driver facial features [J]. Mechanical and Electrical Engineering Technology, 2025, 54 (6): 103-107
10. [10] Zhang K P，Zhang Z P，Li Z F，et al. Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks [J].IEEE Signal Processing Letters，2016，23（10）：1499-1503.
11. [11] Wang Lei, Sun Ruishan Research on Fatigue Monitoring Method for Controllers Based on Facial Feature Recognition [J]. Chinese Journal of Safety Science, 2012, 22 (7): 66-71
12. [12] Gallup A C，Church A M，Pelegrino A J. Yawn Duration Predicts Brain Weight and Cortical Neuron Number in Mammals [J/OL].Biology Letters，2016，12（10）：20160545.https://doi.org/10.1098/rsbl.2016.0545.
13. [13] Gong Shujuan Fatigue driving detection of freight drivers based on facial multi indicators [J]. Logistics Technology, 2020, 39 (9): 74-81, 95
14. [14] Wang Wei, Zhang Xiangwen, Wang Wenhong A safety alert system for detecting driver fatigue status [J]. Sensors and Microsystems, 2022, 41 (6): 82-85
15. [15] Wang Yuanlong, Jing Zhiqiang, Zhang Junwei, etc Research on Lane Line Recognition Methods for Intelligent Vehicles [J]. Journal of Chongqing University of Technology (Natural Science), 2022, 36 (2): 1-10
16. [16] Hu Fengsong, Cheng Zhekun, Xu Qingyun, etc Research on Fatigue Driving State Recognition Method Based on Multi Feature Fusion [J]. Journal of Hunan University (Natural Science Edition), 2022, 49 (4): 100-107
17. [17] Cai Chuangchuang, Liu Qinghua Research on Fatigue Driving Detection Based on Multi Feature Data Fusion [J]. Journal of Jiangsu University of Science and Technology (Natural Science Edition), 2023, 37 (5): 52-57

Funding

Teaching Reform and Innovation Project of Shanxi Datong University (XJG2023262); Applied Basic Research Project of Datong Science and Technology Bureau (2024080); Shanxi Province Higher Education Teaching Reform and Innovation Project (J20241141)

About the authors

Wu Wei (1980-), male, Han ethnicity, Shanxi Datong, lecturer, master's student, research interests: network security, artificial intelligence, computer education; Li Jiajian (2000-), male, Han ethnicity, from Changzhi, Shanxi Province, is currently pursuing a master's degree in recommendation systems.