Abstract—This paper introduces a system of real-time Traffic Light (TL) recognition, which is an essential element for unmanned autonomous vehicles when driving through urban city. The system of TL recognition is an integrating system that fuses a binary processing method, a network model method, and an ITS information. By using two algorithms that processes results through a mono camera, we enhance our recognition accuracy when ITS information is not confirmed properly. We evaluated the individual and integrated TL recognition performance of the system in actual testbed to ensure that our system is satisfied the performance for our autonomous driving scenario. The result of the experiment met our autonomous driving scenario conditions.
Index Terms—autonomous driving, traffic light recognition, ITS, image processing, deep learning
Cite: Mun-Kyu Lee, Jeong-Won Pyo, Sang-Hyeon Bae, Sung-Hyeon Joo, and Tae-Yong Kuc, "Traffic Light Recognition for Autonomous Driving Vehicle: Using Mono Camera and ITS," Journal of Image and Graphics, Vol. 10, No. 3, pp. 102-108, September 2022.
Copyright © 2022 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.
Copyright © 2012-2022 Journal of Image and Graphics, All Rights Reserved