Traffic Sign Detection: Appropriate Data Augmentation Method from the Perspective of Frequency Domain

This study introduces a challenge faced by CNN in the task of traffic sign detection: how to achieve robustness to distributional shift. At present, all kinds of CNN models rely on strong data augmentation methods to enrich samples and achieve robustness, such as Mosaic and Mixup. In this study, we...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Mathematical problems in engineering 2022-12, Vol.2022, p.1-11
Hauptverfasser:	Li, Qingchuan, Zheng, Jiangxing, Tan, Wenfeng, Wang, Xingshu, Zhao, Yingwei
Format:	Artikel
Sprache:	eng
Schlagworte:	Accuracy Data augmentation Datasets Deep learning Experiments Fourier analysis Fourier transforms Frequencies Methods Robustness Traffic control Traffic signs
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!