The detection of alcohol intoxication using electrooculography signals from smart glasses and machine learning techniques

Doniec, Rafał J.; Piaseczna, Natalia; Duraj, Konrad; Sieciński, Szymon; Irshad, Muhammad Tausif; Karpiel, Ilona; Urzeniczok, Mirella; Huang, Xinyu; Piet, Artur; Nisar, Muhammad Adeel; Grzegorzek, Marcin

doi:10.1016/j.sasc.2024.200078

Cited by 2 publications

(1 citation statement)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite the advancements in intelligent transportation systems, drunk driving remains one of the major global traffic safety concerns, significantly contributing to road fatalities. Traditional methods of detecting drunk driving, like breathalyzers or field sobriety tests, are not always accurate or reliable, and drivers may manipulate the results Doniec et al (2024); Greer et al (2023). Consequently, there has been a growing interest in using sensor-based technologies to detect drunk driving.…”

Section: Introductionmentioning

confidence: 99%

A semi-supervised anomaly detection strategy for drunk driving detection: a feasibility study

Harrou,

Kini,

Madakyaru

et al. 2024

Front. Sens.

View full text Add to dashboard Cite

Drunk driving poses a significant threat to road safety, necessitating effective detection methods to enhance preventive measures and ensure the well-being of road users. Recognizing the critical importance of identifying drunk driving incidents for public safety, this paper introduces an effective semi-supervised anomaly detection strategy. The proposed strategy integrates three key elements: Independent Component Analysis (ICA), Kantorovitch distance (KD), and double Exponentially Weighted Moving Average (DEWMA). ICA is used to handle non-gaussian and multivariate data, while KD is used to measure the dissimilarity between normal and abnormal events based on ICA features. The DEWMA is applied to KD charting statistics to detect changes in data and uses a nonparametric threshold to improve sensitivity. The primary advantage of this approach is its ability to perform anomaly detection without requiring labeled data. The study also used XGBoost for the later calculation of the SHAP (SHapley Additive exPlanations) values to identify the most important variables for detecting drunk driving behavior. The approach was evaluated using publicly available data from gas and temperature sensors, as well as digital cameras. The results showed that the proposed approach achieved an F1-score of 98% in detecting the driver’s drunk status, outperforming conventional PCA-based and ICA-based methods.

show abstract

Section: Introductionmentioning

confidence: 99%

A semi-supervised anomaly detection strategy for drunk driving detection: a feasibility study

Harrou,

Kini,

Madakyaru

et al. 2024

Front. Sens.

View full text Add to dashboard Cite

show abstract

Driving Reality vs. Simulator: Data Distinctions

Piaseczna,

Doniec,

Sieciński

et al. 2024

Electronics

View full text Add to dashboard Cite

As the automotive industry undergoes a phase of rapid transformation driven by technological advancements, the integration of driving simulators stands out as an important tool for research and development. The usage of such simulators offers a controlled environment for studying driver behavior; the alignment of data, however, remains a complex aspect that warrants a thorough investigation. This research investigates driver state classification using a dataset obtained from real-road and simulated conditions, recorded through JINS MEME ES_R smart glasses. The dataset encompasses electrooculography signals, with a focus on standardizing and processing the data for subsequent analysis. For this purpose, we used a recurrent neural network model, which yielded a high accuracy on the testing dataset (86.5%). The findings of this study indicate that the proposed methodology could be used in real scenarios and that it could be used for the development of intelligent transportation systems and driver monitoring technology.

show abstract

The detection of alcohol intoxication using electrooculography signals from smart glasses and machine learning techniques

Cited by 2 publications

References 47 publications

A semi-supervised anomaly detection strategy for drunk driving detection: a feasibility study

A semi-supervised anomaly detection strategy for drunk driving detection: a feasibility study

Driving Reality vs. Simulator: Data Distinctions

Contact Info

Product

Resources

About