Vision system warns of driver drowsiness and distraction

Researchers at Universidad Carlos III of Madrid (UC3M; Madrid, Spain) have developed the latest version of a driving assistance system that monitors the driver’s attention level and helps to avoid accidents caused by drowsiness or distractions.

The aim of this system is to supervise and control the driver’s attention level by detecting possible distractions or signs related to drowsiness. “The information comes to the system through a video camera, located on the dashboard of the car,” says professor José María Armingol, who coordinates this research along with Arturo de la Escalera from the Department of Systems Engineering and Automation of UC3M.

The system measures eye blinking speed, and its degree of openness, and where the driver is looking in order to detect potentially dangerous situations. If the system detects that the driver is getting drowsy or that he or she is distracted from the road, the driver is warned by an audible signal to alert him or her to the dangerous situation.

This advanced driving alert system (ADAS) can be applied to practically any type of driver. “It would be logical to start with drivers who spend a lot of time at the wheel, such as truck drivers or bus drivers, and later refine the system to apply it to all types of drivers. Among the possible improvements to the system would be to have it register and evaluate other types of gestures, such as a drooping head or the number of yawns,” says Marco Javier Flores, another of the authors of the research study.

The results of the study have been published recently in the digital version of the Journal of Intelligent and Robotics System.

This new electronic co-pilot is found in the IVVI (Intelligent Vehicle based on Visual Information), a real car that has become a platform for research and experimentation for professors and students at the university. The aim of researchers developing this intelligent vehicle at the Intelligent Systems Laboratory at UC3M is to be able to capture and interpret all of the information on the road when we drive.

The laboratory has also developed a system that detects and classifies the lines on the road to assess the direction the vehicle is headed and thus avoid going off the road. The system can also detect people or obstacles on the road.

Another module of the IVVI integrates a system for traffic sign recognition, which could contribute to improving the state of the roads. This invention detects danger and prohibition signs automatically according to their shape and color using search algorithms. The researchers say that there is a wide range of applications once the system is designed, such as the automatic inspection of traffic signs, based on color, shape, position, and size.