Breathalyzers, step aside. Artificial intelligence is here with a much simpler and noninvasive way to determine if a driver has been imbibing. The model Jack of Many Faces can determine if someone is drunk, tired or angry just by looking at a video feed of their face.
This model was developed by researchers at Edith Cowan University (ECU) to identify when someone may be dangerous behind the wheel by analyzing their facial expressions. It can identify blood alcohol concentration with 90 percent accuracy and is able to classify someone’s level of intoxication as sober, moderate or severe. Its accuracy with noticing drowsiness is even higher at 95 percent.
Add in its ability to register a person’s anger, and it can give a holistic picture of a driver’s current state. It even works in the dark by combining regular video with infrared. “This algorithm is smart, because it can tell the difference between whether a driver is sleepy, just making a facial expression, or affected by alcohol. By separating these factors, it can better understand the driver’s real physical state,” said Syed Zulqarnain Gilani of ECU’s Centre of AI and Machine Learning.
