A camera-based system that can identify drunk drivers just by analysing their faces has been developed by researchers at Edith Cowan University (ECU).

A series of videos were taken of people in a driving simulation who were in various levels of alcohol intoxication – sober, low intoxication and severely intoxicated.

The team developed a machine learning system that uses discernible cues from standard RGB (red, green and blue) videos of the drivers’ faces to gauge the degree of alcohol-related impairment. These included facial features, gaze direction and head position.

“Our system detects varying levels of alcohol intoxication impairment, with an overall accuracy of 75% for the three-level classification,” ECU PhD student Ensiyeh Keshtkaran said.

“This not only benefits vehicles equipped with driver monitoring systems and eye-tracking technologies, but also has the potential to extend to smartphones, making alcohol intoxication detection more effective.

“Our system has the capability to...