Retinex-Based Traffic Sign Recognition for Robust Performance Under Low-Visibility Conditions
International Journal of Advanced Computer Science and Applications · 2026 · 17 (10)
DOI: 10.14569/ijacsa.2026.0171003 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Computer Science and Applications, 2026. Verfasst von Jiyoun Jung und 1 weiteren Autor:innen.
Abstract (Originalsprache)
With the advancement of autonomous vehicles and Advanced Driver Assistance Systems (ADAS), Traffic Sign Recognition (TSR) systems have become increasingly important. However, adverse weather conditions such as fog, rain, and low illumination significantly degrade model performance, leading to missed detections and false positives. To address this challenge, we propose an enhanced TSR model built upon RetinaNet and augmented with a Retinex-based image enhancement algorithm. The model adopts a decoupled detection-classification architecture to improve both accuracy and efficiency. Experiments on the German Traffic Sign Detection Benchmark (GTSDB) and German Traffic Sign Recognition Benchmark (GTSRB) demonstrate that the proposed method partially mitigates detection errors occurring under challenging visibility conditions. For challenging test cases satisfying the Retinex application criterion, the Auto MSR-augmented training configuration improved recall and mAP by approximately four percentage points compared with the original-only baseline. This study highlights the potential of Retinex-based enhancement for improving TSR robustness and contributes to the advancement of intelligent transportation systems.
Zitieren
APA
Jung, J., & Chang, K. (2026). Retinex-Based Traffic Sign Recognition for Robust Performance Under Low-Visibility Conditions. International Journal of Advanced Computer Science and Applications, 17(10). https://doi.org/10.14569/ijacsa.2026.0171003
Vancouver
Jung J, Chang K. Retinex-Based Traffic Sign Recognition for Robust Performance Under Low-Visibility Conditions. International Journal of Advanced Computer Science and Applications. 2026;17(10). doi:10.14569/ijacsa.2026.0171003
BibTeX
@article{Jung2026,
author = {Jung, Jiyoun and Chang, Kyuchang},
title = {Retinex-Based Traffic Sign Recognition for Robust Performance Under Low-Visibility Conditions},
journal = {International Journal of Advanced Computer Science and Applications},
year = {2026},
volume = {17},
number = {10},
doi = {10.14569/ijacsa.2026.0171003}
}
EUVIMED