Researchers at Edith Cowan University in Australia have created a new vehicle camera technology capable of detecting drink driving. According to the article he published IEEE Explore,The system uses image data from standard RGB cameras to detect ,facial signals.
To develop the innovation, 60 volunteers used an indoor driving simulator while a traditional RGB video camera recorded their faces. They were divided into groups with different levels of intoxication: sober, low and severe.
Using a machine learning system, the signals took into account facial features, gaze direction and head position, for example. The technology was able to classify alcohol poisoning with an overall accuracy of 75% across the three levels.
For the researchers, the system could use versions of cameras that face the driver, such as those on the dashboard. That way they could prevent the car from starting if it detects intoxication, unlike methods that rely on long drives to identify problems.
“Our system has the ability to identify levels of intoxication at the beginning of a trip, allowing for the potential prevention of impaired drivers on the road,” Ansia Kashtkaran, a doctoral student at Edith Cowan University, explained in a statement.
Dr. Syed Zulqarnain Jilani, a senior lecturer at the university, adds that the sugar detection camera is the first camera to use a standard RGB camera to detect levels of alcohol poisoning based on signals on the driver’s face.
“The next step in our research is to determine the image resolution needed to use this algorithm. If low-resolution videos are sufficient, this technology could potentially be used by surveillance cameras installed on roads, and law enforcement agencies could use it to prevent driving.” [embriagado]”, he concluded.
It is worth noting that in 2021, Europe was already investing in an immobilizer breathalyzer. An “immobilizer” device connected to the vehicle’s ignition system turns off the engine when it detects that the driver may be under the influence of alcohol. Read more from Brazil chalk.
