Source: Radio New Zealand
New facial recognition technology is being worked on that could detect if someone is driving drunk with just a three second video (file image). 123RF
New facial recognition technology is being worked on that could detect if someone is driving drunk with just a three second video.
Researchers in Australia have been working on the project using artificial intelligence for about two years.
The aim is for it be able to detect whether a person is driving drunk or likely to be a dangerous driver because they are angry or fatigued.
Dr Zulqarnain Gilani from Perth’s Edith Cowan University told First Up the algorithm that’s being developed uses a three to 10 second video of a person to see if they are drunk or fatigued, using their expression.
The technology can also determine a person’s blood alcohol level, Dr Gilani said.
“The algorithm currently can detect five expressions, whether a person is happy, sad, angry or showing disgust, whether they are tired or not tired, or fatigues or not fatigued, and their blood alcohol level as well.”
Through testing, Gilani said videos of people driving a simulator in three different intoxication states with differing blood alcohol levels has been used.
The current technology has a 93 percent accuracy level, he said.
Gilani said it was important that AI used be tested thoroughly on all ethnicities and different conditions.
The current algorithm has been tested on a small cohort of 65 – which was a proof of concept test, he said.
The next steps were to collect more and diverse data if they were to implement this in real life.
Asked how the technology could determine mood, Gilani said it all stemmed around psychology.
“Psychology literature tells us that humans display different, either expressions or psychological states, and their faces show that.
“For example, they say that if somebody is drunk, they blink really fast. And the time for which they close their eyes slows down, so they close it for more time.”
They also suffer hot flushes, he said.
“Whereas if someone is tired, their eyes are droopy. Now the interesting thing is that if somebody is very fatigued and someone is intoxicated, they show almost the same sort of behaviour.”
There were two practical scenarios that the researchers saw for implementing this in real life.
Gilani said the first was to have roadside cameras with the technology which could pick up someone who was driving in an impaired condition and somehow, flag it.
“This is a work in progress. How do we do that and how do we flag it and how do we warn the driver?”
The other was to have the technology inside a person’s car. Gilani said many cars these days have an electric ignition. If a camera facing the driver had the technology and detected a person was impaired, the car wouldn’t start.
Gilani said the project required funds.
“We are actively working with different collaborators, partners and also applying for different fundings so that we can collect more data and make this thing practical.”
Sign up for Ngā Pitopito Kōrero, a daily newsletter curated by our editors and delivered straight to your inbox every weekday.
– Published by EveningReport.nz and AsiaPacificReport.nz, see: MIL OSI in partnership with Radio New Zealand
