Search

A Review of Fatigue Driving Detection Methods Based on Drivers' Facial Features

Author(s)

Sirui Min*

Affiliation(s)

Chang'an Dublin International College of Transportation at Chang'an University, Chang'an University, Xi'an, Shaanxi, China *Corresponding Author

Abstract

Research in the field of fatigue driving detection technology is undergoing a paradigm shift towards multimodal sensing and intelligent decision-making. At this stage, the mainstream methods include three categories: physiological signal monitoring, driving behavior analysis, and facial feature recognition. Among these, the facial feature-based detecting method has become the focus of research due to its non-contact and low-cost advantages. Despite the notable advancements witnessed in this domain, its potential remains constrained by several limitations. These include a paucity of dataset diversity, constrained model generalization capabilities, and an inability to effectively adapt to complex driving environments. In order to enhance the robustness and practicality of the detection system, future research should be initiated with multimodal information fusion, personalized model training, and edge computing deployment. In this paper, the methods and operations of traditional image processing and deep learning techniques are systematically examined. The existing research results are compared and analyzed, and the aim is to provide theoretical support and practical guidance for the optimization of fatigue driving detection technology.

Keywords

Fatigue Driving Detection; Deep Learning; Facial Features

References

[1] Zheng, H. (2024). Research on driver fatigue detection algorithm based on deep learning. Information Technology and Informatization, (12), 49–52. [2] Ding, F., Qin, Y., Zhang, L., & Lü, H. (2024). Driver fatigue state detection method based on multi-feature empirical fusion. Foreign Electronic Measurement Technology, 43(09), 190–196. https://doi.org/10.19652/j.cnki.femt.2406164 [3] Liu, X., Wen, L., Cheng, K., Chen, B., Zhang, Y., & Yu, J. (2024). Driver fatigue detection using improved multi-task cascaded convolutional neural network. Journal of Yibin University, 24(12), 7–11, 68. https://doi.org/10.19504/j.cnki.issn1671-5365.2024.12.02 [4] Xu, J., Tian, J., Liu, X., & Gong, L. (2024). Fatigue driving detection method based on attention mechanism improvement. Sensors and Microsystems, 43(04), 115–118. https://doi.org/10.13873/J.1000-9787(2024)04-0115-04 [5] Zhang, Z., Li, L., Song, J., Fu, H., & Chen, Y. (2024). Research on driver fatigue detection based on deep learning. Electromechanical Engineering Technology, 53(01), 193–197. [6] Jin, Y., Lu, Z., Wang, R., & Liang, C. (2024). Fatigue driving detection based on improved YOLOv5s. Journal of Beihua University (Natural Science Edition), 25(02), 255–261. [7] Zhang, T. (2023). Research on driver fatigue detection based on facial and body behavior features (Master’s thesis, Lanzhou Jiaotong University). https://doi.org/10.27205/d.cnki.gltec.2023.001004 [8] Hu, X., & Huang, B. (2021). Fatigue driving detection method based on facial feature analysis. Science Technology and Engineering, 21(04), 1629–1636. [9] Xiong, Q., Lin, J., & Yue, W. (2018). Fatigue driving state detection method based on deep learning. Control and Information Technology, (06), 91–95. https://doi.org/10.13889/j.issn.2096-5427.2018.06.400 [10] Guo, H., Wang, N., & Guo, H. (2018). Research on fatigue driving warning system based on multiple facial features. Journal on Communications, 39(S1), 22–29. [11] Zou, X., Wang, S., Zhao, W., Zhao, X., & Li, T. (2017). Fatigue driving detection based on eye and mouth state recognition. Journal of Jilin University (Information Science Edition), 35(02), 204–211. https://doi.org/10.19292/j.cnki.jdxxp.2017.02.015 [12] Li, Y., Zhang, C., & Meng, L. (2014). Research on fatigue driving detection system for train drivers based on head pose features. Journal of Transport Information and Safety, 32(05), 114–119.