A Welding Defect Identification Approach in X-ray Images Based on Deep Convolutional Neural Networks

Wang, Yuting; Shi, Fanhuai; Tong, Xuefeng

doi:10.1007/978-3-030-26766-7_6

Cited by 17 publications

(17 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…e mini batch size was taken as 16, the maximum number of epochs was 10, and the learning rate was 0.001 and in general the other parameters were taken to be the same as the proposed structure and the details of each network are replaced as described in subsection 6.1. e ranking performance of each CNN is summarized in Table 5 which illustrates performance comparison for all pretrained models using all chosen performance metrics. e tabular results show that the pretrained AlexNet model reaches the best results followed by VGG- [16][17][18][19] e performance measures presented in Table 5 indicate that the DCNNs models VGG-16, VGG-19, and GoogLeNet proved to be stellar by attaining 95%, 97.8%, and 99.3% accuracy. Moreover, ResNet50 and ResNet101 have accomplished 100% but with more computation time compared to our proposed model.…”

Section: Methodology Of Transfer Learning In Dnns Modelsmentioning

confidence: 99%

“…ere are prominent results for many applications as the visual tasks [20]; also all fields of weld defects detection [15][16][17][18]30] are well investigated. AlexNet [21] is a developed CNN [20] for ImageNet 2012 and the challenge of large-scale visual recognition (ILSVRC-2012).…”

Section: Overview Of Cnn Structure and Detailed Networkmentioning

confidence: 99%

“…Leo et al [16] trained a VGG16 pretrained network for handed cut regions of weld defect rather than the entire image. A similar topic is proposed in [17]. Wang et al proposed a pretrained network based on RetinaNet deep learning procedure to detect various types in a radiographic dataset of welding defects.…”

mentioning

confidence: 97%

See 2 more Smart Citations

Deep Learning Technology for Weld Defects Classification Based on Transfer Learning and Activation Features

Ajmi

Zapata

Zaafouri

et al. 2020

Advances in Materials Science and Engineering

View full text Add to dashboard Cite

Weld defects detection using X-ray images is an effective method of nondestructive testing. Conventionally, this work is based on qualified human experts, although it requires their personal intervention for the extraction and classification of heterogeneity. Many approaches have been done using machine learning (ML) and image processing tools to solve those tasks. Although the detection and classification have been enhanced with regard to the problems of low contrast and poor quality, their result is still unsatisfying. Unlike the previous research based on ML, this paper proposes a novel classification method based on deep learning network. In this work, an original approach based on the use of the pretrained network AlexNet architecture aims at the classification of the shortcomings of welds and the increase of the correct recognition in our dataset. Transfer learning is used as methodology with the pretrained AlexNet model. For deep learning applications, a large amount of X-ray images is required, but there are few datasets of pipeline welding defects. For this, we have enhanced our dataset focusing on two types of defects and augmented using data augmentation (random image transformations over data such as translation and reflection). Finally, a fine-tuning technique is applied to classify the welding images and is compared to the deep convolutional activation features (DCFA) and several pretrained DCNN models, namely, VGG-16, VGG-19, ResNet50, ResNet101, and GoogLeNet. The main objective of this work is to explore the capacity of AlexNet and different pretrained architecture with transfer learning for the classification of X-ray images. The accuracy achieved with our model is thoroughly presented. The experimental results obtained on the weld dataset with our proposed model are validated using GDXray database. The results obtained also in the validation test set are compared to the others offered by DCNN models, which show a best performance in less time. This can be seen as evidence of the strength of our proposed classification model.

show abstract

Section: Methodology Of Transfer Learning In Dnns Modelsmentioning

confidence: 99%

Section: Overview Of Cnn Structure and Detailed Networkmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning Technology for Weld Defects Classification Based on Transfer Learning and Activation Features

Ajmi

Zapata

Zaafouri

et al. 2020

Advances in Materials Science and Engineering

View full text Add to dashboard Cite

show abstract

“…Outras técnicas profundas tratam também desse problema de detecção de defei-tos com redes pré-treinadas, porém, por meio dos feature maps (resultados de camadas intermediárias da rede) para realizar localização de regiões de defeitos [Lin et al 2017, Wang et al 2019. Esta técnicaé conhecida como detecção de objetos, pois destaca as regiões aproximadas dos defeitos de soldagem sobre a radiografia com retângulos englobantes.…”

Section: Trabalhos Relacionadosunclassified

“…Diante da relevância deste tema para os setores supracitados e pelas subjetividades envolvidas no processo de interpretação de juntas soldadas em radiografias por especialistas humanos, a criação de métodos auxiliares e de automatização dessa atividade tem sido foco da comunidade científica [Liao 2003, Nacereddine e Tridi 2005, Baniukiewicz 2014, Wang et al 2019, Duan et al 2019. Assim, esta pesquisa trata de um subcampo de aprendizado de máquina em voga nos anos recentes: o aprendizado profundo, atuando na identificação de defeitos de soldagem de tubulações de petróleo.…”

Section: Introductionunclassified

Filtragem de Dados de Treinamento para Aprendizagem Profunda Aplicada à Inspeção de Juntas Soldadas em Tubulações de Petróleo

Silva¹,

Suyama²,

Silva³

et al. 2020

Anais Do Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2020)

View full text Add to dashboard Cite

A inspeção de radiografias de soldas é uma tarefa essencial para a prevenção de vazamentos em tubulações da indústria petrolífera. A inspeção automática é importante para auxiliar os especialistas em uma tarefa que pode ser influenciada pela fadiga, acuidade visual e experiência. Uma dificuldade da inspeção automática é a disponibilidade de imagens laudadas que possam ser utilizadas para o treinamento de classificadores. O uso de aumento de dados auxilia nesta tarefa mas pode produzir imagens com pouco impacto ou mesmo impacto negativo no treinamento de um classificador. Neste artigo apresenta-se uma método para filtragem de imagens, de forma a favorecer o reaprendizado do classificador, processo este que se dá pela eliminação de imagens com baixo desempenho na base de treino e posterior retreinamento, com impacto positivo para a generalização do modelo de classificação. Os resultados experimentais do método mostram que a filtragem dos dados acelera o processo de aprendizado e produz uma melhora no desempenho da classificação.

show abstract

Images for X-ray Testing

Mery

Pieringer²

2020

Computer Vision for X-Ray Testing

View full text Add to dashboard Cite

A Welding Defect Identification Approach in X-ray Images Based on Deep Convolutional Neural Networks

Cited by 17 publications

References 24 publications

Deep Learning Technology for Weld Defects Classification Based on Transfer Learning and Activation Features

Deep Learning Technology for Weld Defects Classification Based on Transfer Learning and Activation Features

Filtragem de Dados de Treinamento para Aprendizagem Profunda Aplicada à Inspeção de Juntas Soldadas em Tubulações de Petróleo

Images for X-ray Testing

Contact Info

Product

Resources

About