This AI Network Knows If You’re Drunk By Looking At Your Face

Researchers from the Vietnam National University of Ho Chi Minh City have come up with a new artificial intelligence software that could help prevent drunk-driving by detecting if a driver is intoxicated. 

Impaired driving is a major concern around the world, with the Center for Disease Control and Prevention reporting that 29 people in the United States die from alcohol-impaired motor vehicle accidents every day. 

Detailed in the International Journal of Intelligent Information and Database Systems, the network takes a different approach from other detectors, which usually focus on a person’s eyes, head position, or functional state indicators. 

Instead, the team behind the creation posits thermal imaging may be a less ambiguous way to tell if someone’s drunk, and could one day allow police to non-invasively scan drivers’ faces to ensure they don’t get behind the wheel after downing too many beers. 

One of the most important aspects of developing such an algorithm is its degree of accuracy. After all, as Tech Xplore pointed out, a false negative could have terrible consequences if it allows a drunk person to drive, whereas a false positive could hinder others from their commute without reason. 

At the moment, the network boasts an accuracy rate of 93%, though training the AI on larger datasets or on a more diverse population of thermal images could help it become even more accurate in the future. 

[via Tech Xplore, cover image via Iryna Shatilova | Dreamstime.com]