Publication
Adversarial Robustness and Feature Impact Analysis for Driver Drowsiness Detection
| Summary: | 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. |
|---|---|
| Main Authors: | Vitorino, João |
| Other Authors: | Rodrigues, Lourenço; Maia, Eva; Praça, Isabel; Lourenço, André |
| Subject: | Adversarial robustness Explainability Machine learning Heart rate variability Driver drowsiness detection |
| Year: | 2023 |
| Country: | Portugal |
| Document type: | conference output |
| Access type: | open access |
| Associated institution: | Instituto Politécnico do Porto |
| Language: | English |
| Origin: | Repositório Científico do Instituto Politécnico do Porto |
| Summary: | 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. |
|---|