The global statistics of accidents that occurred through the fault of the driver is more than disappointing. Weather conditions and vehicle breakdown cause less than 10% of road traffic accidents (RTAs). The other 90% take place because of carelessness or excessive self-reliance of the driver, often in combination with drunk-driving. The driver safety and driver-friendly systems which are offered by car manufacturers mainly report traffic jams, traffic signs and the distance to the cars ahead, beside or behind. These systems are able to maintain a constant speed on highways and apply emergency braking in case of dangerous proximity to other vehicles. However, such an option is found only in premium-class cars that are beyond the means of many car producers.
In order to improve driving safety, St. Petersburg scientists and programmers have developed an application for a smartphone that can be mounted on the windshield inside the cabin. The system works parallel with navigation (for example, Google Maps) and does not hamper receiving calls and messages. Modern smartphones are multi-functional, boast high processing capacity and have a large number of different built-in sensors. For example, the front camera makes it possible to track the position of the head and the degree of openness of the eyes in order to identify the dangerous signs of sleepiness and low attention as early as possible. The volume of the voice and the redness of the forehead will point to excessively risk-prone, intoxicated or stressed out drivers who will be advised to take a rest to refrain from creating a life-threatening situation on the road.
The scientists have identified the main signs of driver fatigue and turned them into numbers and figures, such as the angle at which the head is bent and the percentage of eye openness. To compile the program, the team used open libraries and databases of computer vision, as well as statistical data which were collected during the experiments of the system utilization by group of volunteers. The application analyzes data from the smartphone’s camera throughout the entire trip and issues warnings that help the driver to cheer up and focus on the driving in case he starts to yawn or “doze off”. In addition, the smartphone registers the speed and the turns, therefore, when the driver looks in the direction other than the one in which he is making the maneuver, the program attracts his attention with sound and light signals thereby preventing an accident. At the moment the application has been downloaded by more than 5000 people all over the world and more than 500 are using it on a regular basis.