Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50

It is difficult to detect and identify natural defects in welded components. To solve this problem, according to the Faraday magneto-optical (MO) effect, a nondestructive testing system for MO imaging, excited by an alternating magnetic field, is established. For the acquired MO images of crack, pit...

Full description

Saved in:
Bibliographic Details
Main Authors: Yanfeng Li, Pengyu Gao, Yongbiao Luo, Xianghan Luo, Chunmei Xu, Jiecheng Chen, Yanxi Zhang, Genxiang Lin, Wei Xu
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/24/23/7649
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850106774749708288
author Yanfeng Li
Pengyu Gao
Yongbiao Luo
Xianghan Luo
Chunmei Xu
Jiecheng Chen
Yanxi Zhang
Genxiang Lin
Wei Xu
author_facet Yanfeng Li
Pengyu Gao
Yongbiao Luo
Xianghan Luo
Chunmei Xu
Jiecheng Chen
Yanxi Zhang
Genxiang Lin
Wei Xu
author_sort Yanfeng Li
collection DOAJ
description It is difficult to detect and identify natural defects in welded components. To solve this problem, according to the Faraday magneto-optical (MO) effect, a nondestructive testing system for MO imaging, excited by an alternating magnetic field, is established. For the acquired MO images of crack, pit, lack of penetration, gas pore, and no defect, Gaussian filtering, bilateral filtering, and median filtering are applied for image preprocessing. The effectiveness of these filtering methods is evaluated using metrics such as peak signal–noise ratio (PSNR) and mean squared error. Principal component analysis (PCA) is employed to extract column vector features from the downsampled defect MO images, which then serve as the input layer for the error backpropagation (BP) neural network model and the support vector machine (SVM) model. These two models can be used for the classification of partial defect MO images, but the recognition accuracy for cracks and gas pores is comparatively low. To further enhance the classification accuracy of natural weld defects, a convolutional neural network (CNN) classification model and a ResNet50 classification model for MO images of natural weld defects are established, and the model parameters are evaluated and optimized. The experimental results show that the overall classification accuracy of the ResNet50 model is 99%. Compared with the PCA-SVM model and CNN model, the overall classification accuracy was increased by 7.4% and 1.8%, and the classification accuracy of gas pore increased by 10% and 4%, respectively, indicating that the ResNet50 model can effectively and accurately classify natural weld defects.
format Article
id doaj-art-10fd3a641dfe4b8380f84bb338091b81
institution OA Journals
issn 1424-8220
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-10fd3a641dfe4b8380f84bb338091b812025-08-20T02:38:45ZengMDPI AGSensors1424-82202024-11-012423764910.3390/s24237649Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50Yanfeng Li0Pengyu Gao1Yongbiao Luo2Xianghan Luo3Chunmei Xu4Jiecheng Chen5Yanxi Zhang6Genxiang Lin7Wei Xu8School of Automobile and Transportation Engineering, Guangdong Polytechnic Normal University, Guangzhou 510632, ChinaGuangdong Provincial Welding Engineering Technology Research Center, Guangdong University of Technology, Guangzhou 510006, ChinaSchool of Automobile and Transportation Engineering, Guangdong Polytechnic Normal University, Guangzhou 510632, ChinaSchool of Automobile and Transportation Engineering, Guangdong Polytechnic Normal University, Guangzhou 510632, ChinaSchool of Computer Science and Technology, Southwest University of Science and Technology, Mianyang 621010, ChinaSchool of Automobile and Transportation Engineering, Guangdong Polytechnic Normal University, Guangzhou 510632, ChinaGuangdong Provincial Welding Engineering Technology Research Center, Guangdong University of Technology, Guangzhou 510006, ChinaSchool of Automobile and Transportation Engineering, Guangdong Polytechnic Normal University, Guangzhou 510632, ChinaSchool of Automobile and Transportation Engineering, Guangdong Polytechnic Normal University, Guangzhou 510632, ChinaIt is difficult to detect and identify natural defects in welded components. To solve this problem, according to the Faraday magneto-optical (MO) effect, a nondestructive testing system for MO imaging, excited by an alternating magnetic field, is established. For the acquired MO images of crack, pit, lack of penetration, gas pore, and no defect, Gaussian filtering, bilateral filtering, and median filtering are applied for image preprocessing. The effectiveness of these filtering methods is evaluated using metrics such as peak signal–noise ratio (PSNR) and mean squared error. Principal component analysis (PCA) is employed to extract column vector features from the downsampled defect MO images, which then serve as the input layer for the error backpropagation (BP) neural network model and the support vector machine (SVM) model. These two models can be used for the classification of partial defect MO images, but the recognition accuracy for cracks and gas pores is comparatively low. To further enhance the classification accuracy of natural weld defects, a convolutional neural network (CNN) classification model and a ResNet50 classification model for MO images of natural weld defects are established, and the model parameters are evaluated and optimized. The experimental results show that the overall classification accuracy of the ResNet50 model is 99%. Compared with the PCA-SVM model and CNN model, the overall classification accuracy was increased by 7.4% and 1.8%, and the classification accuracy of gas pore increased by 10% and 4%, respectively, indicating that the ResNet50 model can effectively and accurately classify natural weld defects.https://www.mdpi.com/1424-8220/24/23/7649magneto-optical imagingnatural weld defectalternating magnetic fieldconvolutional neural networkResNet50
spellingShingle Yanfeng Li
Pengyu Gao
Yongbiao Luo
Xianghan Luo
Chunmei Xu
Jiecheng Chen
Yanxi Zhang
Genxiang Lin
Wei Xu
Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50
Sensors
magneto-optical imaging
natural weld defect
alternating magnetic field
convolutional neural network
ResNet50
title Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50
title_full Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50
title_fullStr Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50
title_full_unstemmed Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50
title_short Automatic Detection and Classification of Natural Weld Defects Using Alternating Magneto-Optical Imaging and ResNet50
title_sort automatic detection and classification of natural weld defects using alternating magneto optical imaging and resnet50
topic magneto-optical imaging
natural weld defect
alternating magnetic field
convolutional neural network
ResNet50
url https://www.mdpi.com/1424-8220/24/23/7649
work_keys_str_mv AT yanfengli automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50
AT pengyugao automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50
AT yongbiaoluo automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50
AT xianghanluo automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50
AT chunmeixu automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50
AT jiechengchen automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50
AT yanxizhang automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50
AT genxianglin automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50
AT weixu automaticdetectionandclassificationofnaturalwelddefectsusingalternatingmagnetoopticalimagingandresnet50