Leveraging driver vehicle and environment interaction: Machine learning using driver monitoring cameras to detect drunk driving

TL;DR

This paper presents a machine learning system that uses driver monitoring cameras to detect drunk driving in real-time, aiming to prevent alcohol-related accidents by identifying intoxication levels during driving.

Contribution

It introduces a novel in-vehicle system leveraging driver cameras to accurately detect drunk driving, validated through a simulator study with promising results.

Findings

Reliable detection of any alcohol influence (AUROC 0.88)

Detection of driving above WHO limit (AUROC 0.79)

First rigorous evaluation of camera-based drunk driving detection

Abstract

Excessive alcohol consumption causes disability and death. Digital interventions are promising means to promote behavioral change and thus prevent alcohol-related harm, especially in critical moments such as driving. This requires real-time information on a person's blood alcohol concentration (BAC). Here, we develop an in-vehicle machine learning system to predict critical BAC levels. Our system leverages driver monitoring cameras mandated in numerous countries worldwide. We evaluate our system with n=30 participants in an interventional simulator study. Our system reliably detects driving under any alcohol influence (area under the receiver operating characteristic curve [AUROC] 0.88) and driving above the WHO recommended limit of 0.05g/dL BAC (AUROC 0.79). Model inspection reveals reliance on pathophysiological effects associated with alcohol consumption. To our knowledge, we are the first to rigorously evaluate the use of driver monitoring cameras for detecting drunk driving. Our results highlight the potential of driver monitoring cameras and enable next-generation drunk driver interaction preventing alcohol-related harm.