A Unified Color and Shape based Algorithm for Traffic Sign Detection System

Abstract

The road safety issue is being in alarming condition due to the rapid increase in roads and vehicles. To solve the scenario, the automotive industry has developed an Advanced Driver Assistance System (ADAS) such as Traffic Sign Detection and Recognition (TSDR), Lane departure warning system, automatic braking system, and so on. Among the driver assistance system, the development of TSDR has the most attention of the researchers and developers due to unsatisfying performance in the detection phase. Variation of illumination level and occlusion by surrounding physical objects are the major concern when developing the TSDR system. In this paper, the combination of color and the shape-based algorithm has been proposed to overcome the issues mentioned. The combination of two different detection techniques which are color thresholding segmentation and edge detection are used to extract traffic signs from the raw image. A detection rate of 85% with a processing time of 0.701s shows the system is a fast system and real-time application.