Traffic Light Detection in Autonomous Vehicles Using Image Processing Methods

Abstract

Despite the great development that the industry has achieved in the field of self-driving cars, However, there are difficulties encountered in detecting traffic lights, in this research, an intelligent system was designed to recognize traffic lights at different times and conditions (day and night). The system is based on digital image processing techniques, which consists of three main stages: image subtraction, traffic signal segmentation, and traffic signal discrimination using two methods, Circular Hough Transform (CH) and morphological operations and comparison between them. The results indicated the efficiency and accuracy of the proposed system in both methods where Traffic lights are detected and the area of each signal is determined over time and the decision is made to detect whether the signal is bright or dark over time.