A robust method for road sign detection and recognition

In this paper a robust method for detecting and recognizing road signs by vision is presented. The detection step is based on a geometrical analysis of the edges extracted from monochromatic images and is able to identify triangular and circular road signs from images of city streets, country roads and highways. A simple recognition system, which validates and classifies the detected road signs, greatly improves the reliability and the robustness of the whole system. Extensive experimentation on real images shows that road signs are usually correctly identified even in cluttered images.