Sept. 8 (UPI) -- Scientists at the University of Waterloo have created a computer algorithm that can accurately determine when drivers are texting while driving.
Distracted drivers are to blame for up to 75 percent of all traffic accidents worldwide, researchers say.
"It has a huge impact on society," Fakhri Karray, a University Research chair and director of the Center for Pattern Analysis and Machine Intelligence, or CPAMI, at Waterloo, said in a press release.
The system uses cameras and artificial intelligence to detect hand movements that deviate from normal driving behavior, such as from texting or talking on the phone, and grades them in terms of safety threats.
The algorithms were trained using machine-learning techniques to recognize actions such as texting, talking on the phone or reaching for something in the backseat or on the floor.
Information received could be used to improve road safety by warning or alerting drivers and, in the case of self-driving cars, could trigger protective measures.
The research builds on previous research at CPAMI on the recognition of signs such as frequent blinking that show drivers are in danger of falling asleep at the wheel. Other signs of distraction include head and face positioning.
"The car could actually take over driving if there was imminent danger, even for a short while, in order to avoid crashes," Karray said.