Circular road sign detection and recognition based on hough transform

In this paper, a new algorithm is proposed to detect and recognize common road signs. Firstly, color segmentation and connected domain analysis are used for region of interest (ROI) detection and then we used the improved two-time Hough transform to separately determinate the circle center and radium of the road sign. Then the Weighted Hausdorff Distance is applied in the recognition of extracted ROI. Compared to the conventional methods, the proposed algorithm has the advantage of small storage, lower time complexity and good robustness towards inclination and occlusion.