In-Car Facial Recognition Could Disable Ignition for Drunk Drivers

Researchers at Australia’s Edith Cowan University have developed a new facial recognition system that aims to prevent intoxicated drivers from operating their vehicles. The system uses regular in-car cameras, typically used as dash cams, to identify signs of intoxication and potentially prevent the car from starting if impairment is detected.

In their study, the researchers tested the system with 60 volunteers using a driving simulator. Each participant drove at three different levels of intoxication: sober, low, and severe. A video camera recorded their faces, and a machine-learning algorithm analyzed facial movements, gaze direction, and head position to identify consistent indicators of intoxication.

The system demonstrated a 75 percent accuracy rate in determining intoxication levels based on these visual characteristics.

The technology builds on existing methods like mini-breathalyzers and pressure-sensing car seats but aims to address their limitations. Unlike systems that track driving patterns and allow a potentially impaired driver to begin operating the vehicle, this facial recognition approach provides a proactive solution by preventing the vehicle from starting.

“This research confirms that it is possible to detect intoxication levels using just a simple camera,” said the project’s lead scientist, Dr. Syed Zulqarnain Gilani. “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 drink driving.”

The system also has the potential for broader applications, such as a smartphone app allowing users to test their sobriety or that of others before driving.

Source: Tomorrow’s World Today

–

July 8, 2024 – by Ali Nassar-Smith