New computer tracking tech that collects data from camera footage to determine if a driver is impaired by alcohol while behind the wheel has been developed by researchers at Edith Cowan University (ECU) in Australia.

The team worked with Mix by Powerfleet to obtain data from alcohol-impaired drivers in a controlled but realistic environment wherein videos were taken of drivers by scientists.

The researchers explained that the drivers represented a range of levels of alcohol intoxication, from sober to low intoxication and severely intoxicated. Those drivers were recorded while driving on a simulator.

The team then presented a machine learning system that uses cues from standard RGB (red, green and blue) videos of the driver's faces — including facial features, gaze direction and head position — to measure the degree of alcohol-related impairment.

The system reportedly detects the different levels of alcohol intoxication impairment with an estimated 75% rate of accuracy.

The researchers suggest that the technology could make alcohol intoxication detection more effective, and that the system is capable of identifying intoxication levels at the start of a drive, there enabling the potential prevention of impaired drivers from being on the road early on.

"This research confirms that it is possible to detect intoxication levels using just a simple camera. The next step in our research is to define the image resolution needed to employ this algorithm. If low resolution videos are proven sufficient, this technology can be employed by surveillance cameras installed on roadside, and law enforcement agencies can use this to prevent [drunk] driving."

An article detailing the technology, “Estimating Blood Alcohol Level Through Facial Features for Driver Impairment Assessment,” was presented at the 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)