Cargando…

Multiscale Traffic Sign Detection Method in Complex Environment Based on YOLOv4

Traffic sign detection is a challenging problem in the field of unmanned driving, particularly important in complex environments. We propose a method, based on the improved You only look once (YOLO) v4, to detect and recognize multiscale traffic signs in complex environments. This method employs an...

Descripción completa

Detalles Bibliográficos
Autores principales: Wang, Yongjie, Bai, Miaoyuan, Wang, Mingzhi, Zhao, Fengfeng, Guo, Jifeng
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Hindawi 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9617714/
https://www.ncbi.nlm.nih.gov/pubmed/36317077
http://dx.doi.org/10.1155/2022/5297605
_version_ 1784820902051971072
author Wang, Yongjie
Bai, Miaoyuan
Wang, Mingzhi
Zhao, Fengfeng
Guo, Jifeng
author_facet Wang, Yongjie
Bai, Miaoyuan
Wang, Mingzhi
Zhao, Fengfeng
Guo, Jifeng
author_sort Wang, Yongjie
collection PubMed
description Traffic sign detection is a challenging problem in the field of unmanned driving, particularly important in complex environments. We propose a method, based on the improved You only look once (YOLO) v4, to detect and recognize multiscale traffic signs in complex environments. This method employs an image preprocessing module that can classify and denoize images of complex environments and then input the images into the improved YOLOv4. We also design an improved feature pyramid structure to replace the original feature pyramid of YOLOv4. This structure uses an adaptive feature fusion module and a multiscale feature transfer mechanism to reduce putative information loss in the feature map generation process and improve the information transfer between deep and shallow features, enhancing the representation ability of feature pyramids. Finally, we use EIOU LOSS and Cluster-NMS to further improve the model performance. The experimental results on the fusion of Tsinghua-Tencent 100 K and our collected dataset show that the proposed method achieves an mAP of 81.78%. Compared to existing methods, our method demonstrates its superiority with regard to traffic sign detection.
format Online
Article
Text
id pubmed-9617714
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Hindawi
record_format MEDLINE/PubMed
spelling pubmed-96177142022-10-30 Multiscale Traffic Sign Detection Method in Complex Environment Based on YOLOv4 Wang, Yongjie Bai, Miaoyuan Wang, Mingzhi Zhao, Fengfeng Guo, Jifeng Comput Intell Neurosci Research Article Traffic sign detection is a challenging problem in the field of unmanned driving, particularly important in complex environments. We propose a method, based on the improved You only look once (YOLO) v4, to detect and recognize multiscale traffic signs in complex environments. This method employs an image preprocessing module that can classify and denoize images of complex environments and then input the images into the improved YOLOv4. We also design an improved feature pyramid structure to replace the original feature pyramid of YOLOv4. This structure uses an adaptive feature fusion module and a multiscale feature transfer mechanism to reduce putative information loss in the feature map generation process and improve the information transfer between deep and shallow features, enhancing the representation ability of feature pyramids. Finally, we use EIOU LOSS and Cluster-NMS to further improve the model performance. The experimental results on the fusion of Tsinghua-Tencent 100 K and our collected dataset show that the proposed method achieves an mAP of 81.78%. Compared to existing methods, our method demonstrates its superiority with regard to traffic sign detection. Hindawi 2022-10-22 /pmc/articles/PMC9617714/ /pubmed/36317077 http://dx.doi.org/10.1155/2022/5297605 Text en Copyright © 2022 Yongjie Wang et al. https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
spellingShingle Research Article
Wang, Yongjie
Bai, Miaoyuan
Wang, Mingzhi
Zhao, Fengfeng
Guo, Jifeng
Multiscale Traffic Sign Detection Method in Complex Environment Based on YOLOv4
title Multiscale Traffic Sign Detection Method in Complex Environment Based on YOLOv4
title_full Multiscale Traffic Sign Detection Method in Complex Environment Based on YOLOv4
title_fullStr Multiscale Traffic Sign Detection Method in Complex Environment Based on YOLOv4
title_full_unstemmed Multiscale Traffic Sign Detection Method in Complex Environment Based on YOLOv4
title_short Multiscale Traffic Sign Detection Method in Complex Environment Based on YOLOv4
title_sort multiscale traffic sign detection method in complex environment based on yolov4
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9617714/
https://www.ncbi.nlm.nih.gov/pubmed/36317077
http://dx.doi.org/10.1155/2022/5297605
work_keys_str_mv AT wangyongjie multiscaletrafficsigndetectionmethodincomplexenvironmentbasedonyolov4
AT baimiaoyuan multiscaletrafficsigndetectionmethodincomplexenvironmentbasedonyolov4
AT wangmingzhi multiscaletrafficsigndetectionmethodincomplexenvironmentbasedonyolov4
AT zhaofengfeng multiscaletrafficsigndetectionmethodincomplexenvironmentbasedonyolov4
AT guojifeng multiscaletrafficsigndetectionmethodincomplexenvironmentbasedonyolov4