Traffic Sign Detection Via Color and Shape-Based Approach

Abstract

Traffic sign detection is one of the basic operations for intelligent transport systems and driver assistance systems. In this study, a new method for traffic sign detection based on image processing was proposed. The performance of the proposed method has been increased by taking into account both color information and geometric shapes of traffic signs in the feature extraction phase. The RGB color space was used in the study. First, noise reduction and morphological processes were made the image formally tractable. After that by dint of the shape-based approach, triangular, circular and square shapes were taken into consideration, shapes were identified by means of connected components and figures without traffic signs were removed from the image. Thus, the remaining objects formed the traffic signs in the image. As a result of the experimental study applied to the data set, it was seen that the traffic signs in the image were detected. © 2019 IEEE.