Traffic signs recognition for detailed digital maps development and driver assistance systems

Resumen

Digital maps are considered as an additional sensor in many of the new ADAS, but these systems usually require a higher level of accuracy and detail of the maps. Among the important information that the maps should contain are the road geometry and traffic signs. In the first case, it is interesting to use accurate and fast methods for measurement. In the paper, a method based on a datalog vehicle is used. Satellite positioning and inertial measurements systems data are combined and dynamic behavior of the vehicle body is corrected measuring the movements of the suspension system. On the other hand, the information provided by traffic signs and route-guidance signs is extremely important for safe and successful driving. An automatic system that is capable of extracting and identifying these signs automatically would help human drivers enormously; navigation would be easier, allowing them to concentrate on driving the vehicle. A Computer Vision System is used to recognize and classify the different families of traffic signs combining it with GPS information to develop detailed and accurate digital maps. This sign recognition can also be used for real time warnings to the driver. Some results of test carried out in real situations are shown.