A Study on Railway Surface Defects Detection Based on Machine Vision

The detection of rail surface defects is an important tool to ensure the safe operation of rail transit. Due to the complex diversity of track surface defect features and the small size of the defect area, it is difficult to obtain satisfying detection results by traditional machine vision methods....

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Tangbo Bai, Jialin Gao, Jianwei Yang, Dechen Yao
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	deep learning rail surface defect detection machine vision YOLOv4 MobileNetV3 Science Q Astrophysics QB460-466 Physics QC1-999
Acceso en línea:	https://doaj.org/article/f0d2f43b37334dc191a8c02bd32f0e10
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:f0d2f43b37334dc191a8c02bd32f0e10
record_format	dspace
spelling	oai:doaj.org-article:f0d2f43b37334dc191a8c02bd32f0e102021-11-25T17:29:39ZA Study on Railway Surface Defects Detection Based on Machine Vision10.3390/e231114371099-4300https://doaj.org/article/f0d2f43b37334dc191a8c02bd32f0e102021-10-01T00:00:00Zhttps://www.mdpi.com/1099-4300/23/11/1437https://doaj.org/toc/1099-4300The detection of rail surface defects is an important tool to ensure the safe operation of rail transit. Due to the complex diversity of track surface defect features and the small size of the defect area, it is difficult to obtain satisfying detection results by traditional machine vision methods. The existing deep learning-based methods have the problems of large model sizes, excessive parameters, low accuracy and slow speed. Therefore, this paper proposes a new method based on an improved YOLOv4 (You Only Look Once, YOLO) for railway surface defect detection. In this method, MobileNetv3 is used as the backbone network of YOLOv4 to extract image features, and at the same time, deep separable convolution is applied on the PANet layer in YOLOv4, which realizes the lightweight network and real-time detection of the railway surface. The test results show that, compared with YOLOv4, the study can reduce the amount of the parameters by 78.04%, speed up the detection by 10.36 frames per second and decrease the model volume by 78%. Compared with other methods, the proposed method can achieve a higher detection accuracy, making it suitable for the fast and accurate detection of railway surface defects.Tangbo BaiJialin GaoJianwei YangDechen YaoMDPI AGarticledeep learningrail surface defect detectionmachine visionYOLOv4MobileNetV3ScienceQAstrophysicsQB460-466PhysicsQC1-999ENEntropy, Vol 23, Iss 1437, p 1437 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	deep learning rail surface defect detection machine vision YOLOv4 MobileNetV3 Science Q Astrophysics QB460-466 Physics QC1-999
spellingShingle	deep learning rail surface defect detection machine vision YOLOv4 MobileNetV3 Science Q Astrophysics QB460-466 Physics QC1-999 Tangbo Bai Jialin Gao Jianwei Yang Dechen Yao A Study on Railway Surface Defects Detection Based on Machine Vision
description	The detection of rail surface defects is an important tool to ensure the safe operation of rail transit. Due to the complex diversity of track surface defect features and the small size of the defect area, it is difficult to obtain satisfying detection results by traditional machine vision methods. The existing deep learning-based methods have the problems of large model sizes, excessive parameters, low accuracy and slow speed. Therefore, this paper proposes a new method based on an improved YOLOv4 (You Only Look Once, YOLO) for railway surface defect detection. In this method, MobileNetv3 is used as the backbone network of YOLOv4 to extract image features, and at the same time, deep separable convolution is applied on the PANet layer in YOLOv4, which realizes the lightweight network and real-time detection of the railway surface. The test results show that, compared with YOLOv4, the study can reduce the amount of the parameters by 78.04%, speed up the detection by 10.36 frames per second and decrease the model volume by 78%. Compared with other methods, the proposed method can achieve a higher detection accuracy, making it suitable for the fast and accurate detection of railway surface defects.
format	article
author	Tangbo Bai Jialin Gao Jianwei Yang Dechen Yao
author_facet	Tangbo Bai Jialin Gao Jianwei Yang Dechen Yao
author_sort	Tangbo Bai
title	A Study on Railway Surface Defects Detection Based on Machine Vision
title_short	A Study on Railway Surface Defects Detection Based on Machine Vision
title_full	A Study on Railway Surface Defects Detection Based on Machine Vision
title_fullStr	A Study on Railway Surface Defects Detection Based on Machine Vision
title_full_unstemmed	A Study on Railway Surface Defects Detection Based on Machine Vision
title_sort	study on railway surface defects detection based on machine vision
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/f0d2f43b37334dc191a8c02bd32f0e10
work_keys_str_mv	AT tangbobai astudyonrailwaysurfacedefectsdetectionbasedonmachinevision AT jialingao astudyonrailwaysurfacedefectsdetectionbasedonmachinevision AT jianweiyang astudyonrailwaysurfacedefectsdetectionbasedonmachinevision AT dechenyao astudyonrailwaysurfacedefectsdetectionbasedonmachinevision AT tangbobai studyonrailwaysurfacedefectsdetectionbasedonmachinevision AT jialingao studyonrailwaysurfacedefectsdetectionbasedonmachinevision AT jianweiyang studyonrailwaysurfacedefectsdetectionbasedonmachinevision AT dechenyao studyonrailwaysurfacedefectsdetectionbasedonmachinevision
_version_	1718412321201586176

A Study on Railway Surface Defects Detection Based on Machine Vision

Ejemplares similares