Abstract
Drowsy driving is a major cause of road accidents, but drivers are dismissive of the impact that fatigue can have on their reaction times. To detect drowsiness before any impairment occurs, a promising strategy is using Machine Learning (ML) to monitor Heart Rate Variability (HRV) signals. This work presents multiple experiments with different HRV time windows and ML models, a feature impact analysis using Shapley Additive Explanations (SHAP), and an adversarial robustness analysis to assess their reliability when processing faulty input data and perturbed HRV signals. The most reliable model was Extreme Gradient Boosting (XGB) and the optimal time window had between 120 and 150 s. Furthermore, the 18 most impactful features were selected and new smaller models were trained, achieving a performance as good as the initial ones. Despite the susceptibility of all models to adversarial attacks, adversarial training enabled them to preserve significantly higher results, so it can be a valuable approach to provide a more robust driver drowsiness detection.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
European Commission, Road safety thematic report – Fatigue, (2021). https://road-safety.transport.ec.europa.eu/statistics-and-analysis/data-and-analysis/thematic-reports_en
Oliveira, L., et al.: Driver drowsiness detection: a comparison between intrusive and non-intrusive signal acquisition methods, In: 7th Eur. Work. Visu. Info. Process., pp. 1–6 (2018)
Chowdhury, A., et al.: Sensor Applications and Physiological Features in Drivers’ Drowsiness Detection: A Review. IEEE Sens. J. (2018). https://doi.org/10.1109/JSEN.2018.2807245
Lourenço, A., et al.: CardioWheel: ECG Biometrics on the Steering Wheel, In: Machine Learning and Knowledge Discovery in Databases, pp. 267–270 (2015)
Fors, C., et al.: Camera-based sleepiness detection: final report of the project SleepEYE, (2011) https://www.academia.edu/34322032/Camera_based_sleepiness_detection_Final_report_of_the_project_SleepEYE
Silveira, C.S., et al.: Importance of subject-dependent classification and imbalanced distributions in driver sleepiness detection in realistic conditions. IET Intell. Transp. Syst. 13(2), 347–355 (2019). https://doi.org/10.1049/iet-its.2018.5284
Rodrigues, L.: Driver Drowsiness Detection with Peripheral Cardiac Signals, Instituto Superior Técnico, (2021) https://fenix.tecnico.ulisboa.pt/cursos/mebiom/dissertacao/ 1128253548922289
Massoz, Q., et al.: The ULg multimodality drowsiness database (called DROZY) and examples of use. In: IEEE Winter Conf. on Appl. of Computer Vision, pp. 1–7 (2016)
Vitorino, J., et al.: Towards Adversarial Realism and Robust Learning for IoT Intrusion Detection and Classification. Ann. Telecommun. (2023). https://doi.org/10.1007/s12243-023-00953-y
Vitorino, J., et al.: Adaptative Perturbation Patterns: Realistic Adversarial Learning for Robust Intrusion Detection, Future Internet, vol. 14(4) (2022) https://doi.org/10.3390/fi14040108
Hearst, M.A., et al.: Support vector machines. IEEE Intell. Syst. their Appl. 13(4), 18–28 (1998). https://doi.org/10.1109/5254.708428
Cover, T., Hart, P.: Nearest neighbor pattern classification. IEEE Trans. Inf. Theory 13(1), 21–27 (1967). https://doi.org/10.1109/TIT.1967.1053964
Chen, T., Guestrin, C.: XGBoost: a scalable tree boosting system In: Proceedings of the ACM Int. Conf. on Knowl. Discov. and Data Min., (2016) https://doi.org/10.1145/2939672.2939785
Ke, G., et al.: LightGBM: a highly efficient gradient boosting decision tree, in Advan. in Neural Info. Process. Sys., 2017, pp. 3147–3155 (2017)
Vitorino, J., et al.: A Comparative Analysis of Machine Learning Techniques for IoT Intrusion Detection. Found. Pract. of Sec. (2022). https://doi.org/10.1007/978-3-031-08147-7_13
Lundberg, S. M., Lee, S.-I.: A Unified Approach to Interpreting Model Predictions, in Advan. in Neural Info. Process. Sys., pp. 4765–4774 (2017)
Acknowledgments
This work was done and funded in the scope of the European Union's Horizon 2020 research and innovation program, under project VALU3S (grant agreement no. 876852). This work has also received funding from UIDP/00760/2020. A publicly available dataset was utilized in this work. The data can be found at: https://hdl.handle.net/2268/191620.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Vitorino, J., Rodrigues, L., Maia, E., Praça, I., Lourenço, A. (2023). Adversarial Robustness and Feature Impact Analysis for Driver Drowsiness Detection. In: Juarez, J.M., Marcos, M., Stiglic, G., Tucker, A. (eds) Artificial Intelligence in Medicine. AIME 2023. Lecture Notes in Computer Science(), vol 13897. Springer, Cham. https://doi.org/10.1007/978-3-031-34344-5_13
Download citation
DOI: https://doi.org/10.1007/978-3-031-34344-5_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-34343-8
Online ISBN: 978-3-031-34344-5
eBook Packages: Computer ScienceComputer Science (R0)