The detection and recognition of traffic signs is an essential component intelligent vehicle perception systems, which use on-board cameras to sense sign information. Unfortunately, issues such as long-tailed distribution, occlusion, deformation greatly decrease the detector’s performance. In this research, YOLOv5 used a single classification detector for localization. Afterwards, we propose hi...

This study introduces a challenge faced by CNN in the task of traffic sign detection: how to achieve robustness distributional shift. At present, all kinds models rely on strong data augmentation methods enrich samples and robustness, such as Mosaic Mixup. In this study, we note that these do not have similar effects combating noise. We explore performance strategies against disturbance differe...

Traffic sign and mile-marker detection and classification are among the important components of highway asset management systems. The significant number of these high-quantity and low-cost assets in US highways can negatively impact on quality of any manual data collection and analysis. To address these challenges, this paper presents an efficient pipeline for video-based detection and classifi...

In this paper, we propose a new pre-processing method for detecting road traffic sign based on visual saliency map model. Since the road traffic sign boards have dominent color contrast against environment, we consider the color opponents information with center surround difference normalization as an input feature extraction, which is effective to reduce noise influence as well as intensify th...