Detection and recognition of road signs

Abstract

In this paper, we present a system for the detection and recognition of road signs based on computer vision. The approach adopted in this work consists of two modules, a detection module which first segments the image using a combination of RGB and HSV colors spaces and then searches for relevant shapes (circles, triangles, squares.) using the Hough transform and corner detection. And a recognition module which uses support vector machines (SVMs) to recognize each type of road signs learned a priori using the SVM classifier with Zernike moments, histogram of gradients (HOG) and local binary patterns (LBP) . Finally, we discuss the performance of the system in terms of detection and robustness.