基于YOLOv8的驾驶员分心状态识别算法研究
Research on Driver Distracted State Recognition Algorithm Based on YOLOv8
DOI: 10.12677/sea.2026.151005, PDF,   
作者: 庞夏君, 刘玉情：广西职业技术学院人工智能学院，广西 南宁
关键词: 分心驾驶；YOLOv8n；注意力机制；小目标检测；Distracted Driving； YOLOv8n； Attention Mechanism； Small Target Detection
摘要: 准确、快速检测分心驾驶行为对于提高道路安全和预防交通事故至关重要。从驾驶员分心驾驶行为特点出发，提出了基于改进YOLOv8n的分心驾驶行为检测算法。首先是引入了简单注意力机制(SimAM)对每个特征图进行自归一化，强调具有有价值信息的特征图，抑制复杂背景中的冗余信息干扰。其次为了提高对小目标物体的检测性能，增加了小目标检测层，使模型能够更准确地检测和定位尺寸较小的目标，提高目标检测系统在复杂场景下的稳定性和鲁棒性。最后在公开和收集到的数据集上进行了验证，模型精度提高了4.2%。分心驾驶检测准确率分别达到90.4%验证了所提模型在辅助驾驶时分心驾驶检测和降低事故风险方面的卓越性能。
Abstract: Accurate and rapid detection of distracted driving behavior is critical to improving road safety and preventing traffic accidents. Based on the characteristics of drivers’ distracted driving behavior, an improved YOLOv8n distracted driving behavior detection algorithm is proposed. Firstly, a simple attention mechanism (SimAM) is introduced to self-normalize each feature map, emphasizing the feature maps with valuable information and suppressing the interference of redundant information in complex background. Secondly, in order to improve the detection performance of small target objects, a small target detection layer is added, so that the model can detect and locate smaller targets more accurately, and improve the stability and robustness of the target detection system in complex scenes. Finally, the accuracy of the model is improved by 4.2% on the published and collected data sets. The accuracy of distracted driving detection reached 90.4% respectively, which verified the excellent performance of the proposed model in detecting distracted driving and reducing accident risk during assisted driving.
文章引用：庞夏君, 刘玉情. 基于YOLOv8的驾驶员分心状态识别算法研究[J]. 软件工程与应用, 2026, 15(1): 38-48. https://doi.org/10.12677/sea.2026.151005

参考文献

[1] 金朝力, 冉黎黎. 2023年全国机动车保有量达4.35亿辆[N]. 北京商报, 2024-01-12(003).
[2] 巩建国, 赵琳娜. 分心驾驶行为演变与防治对策研究[J]. 汽车与安全, 2015(11): 69-72.
[3] Chu, Y., Chowdhury, H.R., Mitul, A.F., Uddin, N., Hossain, M.J. and Aslam, D.M. (2019) Nature of Distracted Driving in Various Physiological Conditions. Applications of Machine Learning, 11139, 11139B. [Google Scholar] [CrossRef] 
[4] Tran, D., Manh Do, H., Sheng, W., Bai, H. and Chowdhary, G. (2018) Real-Time Detection of Distracted Driving Based on Deep Learning. IET Intelligent Transport Systems, 12, 1210-1219. [Google Scholar] [CrossRef] 
[5] 梁昭德. 基于卷积神经网络的驾驶人疲劳检测算法研究[D]: [硕士学位论文]. 广州: 华南理工大学, 2018.
[6] 陈军, 张黎, 周博, 等. 基于级联卷积神经网络的驾驶员分心驾驶行为检测[J]. 科学技术与工程, 2020, 20(14): 5702-5708.
[7] 李涛. 基于深度学习的复杂场景下车载图像目标检测方法研究[D]: [硕士学位论文]. 重庆: 重庆大学, 2018.
[8] 孙玥, 杨国为, 陈雪鑫. 基于计算机视觉的疲劳及注意力检测算法[J]. 计算机与数字工程, 2021, 49(6): 1195-1198.
[9] Du, Y., Liu, X., Yi, Y. and Wei, K. (2023) Optimizing Road Safety: Advancements in Lightweight YOLOv8 Models and GhostC2f Design for Real-Time Distracted Driving Detection. Sensors, 23, Article 8844. [Google Scholar] [CrossRef] [PubMed]
[10] 周飞, 郭杜杜, 王洋, 等. 基于改进YOLOv8的交通监控车辆检测算法[J]. 计算机工程与应用, 2024, 60(6): 110-120.
[11] Ultralytics (2026) YOLOv8 Documentation.
https://docs.ultralytics.com/models/yolov8/
[12] 蒋昂波, 王维维. ReLU激活函数优化研究[J]. 传感器与微系统, 2018, 37(2): 50-52.
[13] 宋鹏飞, 吴云. DenseNet和SeNet融合残差结构的DR分类方法[J]. 计算机应用研究, 2024, 41(3): 928-932+950.
[14] 刘丽伟, 侯德彪, 侯阿临, 等. 基于SimAM-YOLOv4的自动驾驶目标检测算法[J]. 长春工业大学学报, 2022, 43(3): 244-250.
[15] Perez, J., Hyun, K. and Boni, J.A. (2024) Use of Smartphone Apps While Driving: Variations on Driving Performances and Perceived Risks. Accident Analysis & Prevention, 198, Article 107474. [Google Scholar] [CrossRef] [PubMed]
[16] Abbass, M.A.B. and Ban, Y. (2024) Mobilenet-Based Architecture for Distracted Human Driver Detection of Autonomous Cars. Electronics, 13, Article 365. [Google Scholar] [CrossRef] 
[17] Demir, B., Demir, S., Wickens, C.M., et al. (2024) Exploring the Behaviour Change Wheel and the Theoretical Domains Framework in Interventions for Mobile Phone Driver Distraction: A Scoping Review. Accident Analysis and Prevention, 195, Article 107369.
[18] Chen, Z.J., Xie, L., Zhang, J. and Lin, L. (2021) SimAM: A Simple, Parameter-Free Attention Module for Convolutional Neural Networks. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, 19-25 June 2021, 11899-11908.
[19] Leekha, M., Goswami, M., Shah, R.R., Yin, Y. and Zimmermann, R. (2019) Are You Paying Attention? Detecting Distracted Driving in Real-Time. 2019 IEEE Fifth International Conference on Multimedia Big Data (BigMM), Singapore, 11-13 September 2019, 171-180. [Google Scholar] [CrossRef] 
[20] Alzubaidi, L., Zhang, J., Humaidi, A.J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., et al. (2021) Review of Deep Learning: Concepts, CNN Architectures, Challenges, Applications, Future Directions. Journal of Big Data, 8, Article No. 53. [Google Scholar] [CrossRef] [PubMed]
[21] Eraqi, H.M., Abouelnaga, Y., Saad, M.H. and Moustafa, M.N. (2019) Driver Distraction Identification with an Ensemble of Convolutional Neural Networks. Journal of Advanced Transportation, 2019, 1-12. [Google Scholar] [CrossRef] 
[22] Liu, S., Qi, L., Qin, H., Shi, J. and Jia, J. (2018) Path Aggregation Network for Instance Segmentation. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, 18-23 June 2018, 8759-8768. [Google Scholar] [CrossRef] 

为你推荐