Real Time Driving-Aid System for Different Lighting Conditions, on Board a Road Vehicle

In this work a system for traffic-sign detection and classification under different lighting conditions is shown. It is intended for circular and triangular signs. The system is composed of three stages: first, detection, using the Hough transform for lines and circumferences from the information of the edges of the image instead of the whole image information; second, tracking, making use of a Kalman filter, which provides the system with memory, and third, classification, using a neural network. Some results are presented, obtained with real images recorded by only one camera placed on board a conventional vehicle, in sunny, cloudy and rainy days, and also at night, in order to show the reliability and robustness of the system with different light conditions. The average processing time is 30 ms per frame, what makes this work a good approach to work in real time conditions