Deep convolutional neural network for weld defect classification in radiographic images
The quality of welds is critical to the safety of structures in construction, so early detection of irregularities is crucial. Advances in machine vision inspection technologies, such as deep learning models, have improved the detection of weld defects. This paper presents a new CNN model based on R...
- Autores:
-
Palma Ramírez, Dayana
Ross Veitía, Bárbara D.
Font Ariosa, Pablo
Espinel Hernández, Alejandro
Sánchez Roca, Ángel
Carvajar Fals, Hipólito
Nuñez Álvarez, José R.
Hernández Herrera, Hernan
- Tipo de recurso:
- Article of investigation
- Fecha de publicación:
- 2024
- Institución:
- Corporación Universidad de la Costa
- Repositorio:
- REDICUC - Repositorio CUC
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.cuc.edu.co:11323/13481
- Acceso en línea:
- https://hdl.handle.net/11323/13481
https://repositorio.cuc.edu.co/
- Palabra clave:
- Radiographic testing
Classification
Weld defects
CNNs
Transfer learning
- Rights
- openAccess
- License
- Atribución 4.0 Internacional (CC BY 4.0)
| id |
RCUC2_db9b2cbe336ec76644c913278d85d3ae |
|---|---|
| oai_identifier_str |
oai:repositorio.cuc.edu.co:11323/13481 |
| network_acronym_str |
RCUC2 |
| network_name_str |
REDICUC - Repositorio CUC |
| repository_id_str |
|
| dc.title.eng.fl_str_mv |
Deep convolutional neural network for weld defect classification in radiographic images |
| title |
Deep convolutional neural network for weld defect classification in radiographic images |
| spellingShingle |
Deep convolutional neural network for weld defect classification in radiographic images Radiographic testing Classification Weld defects CNNs Transfer learning |
| title_short |
Deep convolutional neural network for weld defect classification in radiographic images |
| title_full |
Deep convolutional neural network for weld defect classification in radiographic images |
| title_fullStr |
Deep convolutional neural network for weld defect classification in radiographic images |
| title_full_unstemmed |
Deep convolutional neural network for weld defect classification in radiographic images |
| title_sort |
Deep convolutional neural network for weld defect classification in radiographic images |
| dc.creator.fl_str_mv |
Palma Ramírez, Dayana Ross Veitía, Bárbara D. Font Ariosa, Pablo Espinel Hernández, Alejandro Sánchez Roca, Ángel Carvajar Fals, Hipólito Nuñez Álvarez, José R. Hernández Herrera, Hernan |
| dc.contributor.author.none.fl_str_mv |
Palma Ramírez, Dayana Ross Veitía, Bárbara D. Font Ariosa, Pablo Espinel Hernández, Alejandro Sánchez Roca, Ángel Carvajar Fals, Hipólito Nuñez Álvarez, José R. Hernández Herrera, Hernan |
| dc.subject.proposal.eng.fl_str_mv |
Radiographic testing Classification Weld defects CNNs Transfer learning |
| topic |
Radiographic testing Classification Weld defects CNNs Transfer learning |
| description |
The quality of welds is critical to the safety of structures in construction, so early detection of irregularities is crucial. Advances in machine vision inspection technologies, such as deep learning models, have improved the detection of weld defects. This paper presents a new CNN model based on ResNet50 to classify four types of weld defects in radiographic images: crack, pore, non-penetration, and no defect. Stratified cross-validation, data augmentation, and regularization were used to improve generalization and avoid over-fitting. The model was tested on three datasets, RIAWELC, GDXray, and a private dataset of low image quality, obtaining an accuracy of 98.75 %, 90.255 %, and 75.83 %, respectively. The model proposed in this paper achieves high accuracies on different datasets and constitutes a valuable tool to improve the efficiency and effectiveness of quality control processes in the welding industry. Moreover, experimental tests show that the proposed approach performs well on even low-resolution images. |
| publishDate |
2024 |
| dc.date.accessioned.none.fl_str_mv |
2024-10-23T12:40:23Z |
| dc.date.available.none.fl_str_mv |
2024-10-23T12:40:23Z |
| dc.date.issued.none.fl_str_mv |
2024-05-15 |
| dc.type.none.fl_str_mv |
Artículo de revista |
| dc.type.coar.none.fl_str_mv |
http://purl.org/coar/resource_type/c_2df8fbb1 |
| dc.type.content.none.fl_str_mv |
Text |
| dc.type.driver.none.fl_str_mv |
info:eu-repo/semantics/article |
| dc.type.redcol.none.fl_str_mv |
http://purl.org/redcol/resource_type/ART |
| dc.type.version.none.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
| dc.type.coarversion.none.fl_str_mv |
http://purl.org/coar/version/c_970fb48d4fbd8a85 |
| format |
http://purl.org/coar/resource_type/c_2df8fbb1 |
| status_str |
publishedVersion |
| dc.identifier.citation.none.fl_str_mv |
Dayana Palma-Ramírez, Bárbara D. Ross-Veitía, Pablo Font-Ariosa, Alejandro Espinel-Hernández, Angel Sanchez-Roca, Hipólito Carvajal-Fals, José R. Nuñez-Alvarez, Hernan Hernández-Herrera, Deep convolutional neural network for weld defect classification in radiographic images, Heliyon, Volume 10, Issue 9, 2024, e30590, ISSN 2405-8440, https://doi.org/10.1016/j.heliyon.2024.e30590. |
| dc.identifier.uri.none.fl_str_mv |
https://hdl.handle.net/11323/13481 |
| dc.identifier.doi.none.fl_str_mv |
10.1016/j.heliyon.2024.e30590 |
| dc.identifier.eissn.none.fl_str_mv |
2405-8440 |
| dc.identifier.instname.none.fl_str_mv |
Corporación Universidad de la Costa |
| dc.identifier.reponame.none.fl_str_mv |
REDICUC - Repositorio CUC |
| dc.identifier.repourl.none.fl_str_mv |
https://repositorio.cuc.edu.co/ |
| identifier_str_mv |
Dayana Palma-Ramírez, Bárbara D. Ross-Veitía, Pablo Font-Ariosa, Alejandro Espinel-Hernández, Angel Sanchez-Roca, Hipólito Carvajal-Fals, José R. Nuñez-Alvarez, Hernan Hernández-Herrera, Deep convolutional neural network for weld defect classification in radiographic images, Heliyon, Volume 10, Issue 9, 2024, e30590, ISSN 2405-8440, https://doi.org/10.1016/j.heliyon.2024.e30590. 10.1016/j.heliyon.2024.e30590 2405-8440 Corporación Universidad de la Costa REDICUC - Repositorio CUC |
| url |
https://hdl.handle.net/11323/13481 https://repositorio.cuc.edu.co/ |
| dc.language.iso.none.fl_str_mv |
eng |
| language |
eng |
| dc.relation.ispartofjournal.none.fl_str_mv |
Heliyon |
| dc.relation.references.none.fl_str_mv |
T.J. McPheron, R.M. Stwalley, Engineering Challenges Associated with Welding Field Repairs, Engineering Principles-Welding and Residual Stresses, 2022, https://doi.org/10.5772/intechopen.104263. M. P´erez de la Parte, et al., Effect of zinc coating on delay nugget formation in dissimilar DP600-AISI304 welded joints obtained by the resistance spot welding process, Int. J. Adv. Des. Manuf. Technol. 120 (2022) 1877–1887, https://doi.org/10.1007/s00170-022-08849-2. D. Varshney, K. Kumar, Application and use of different aluminium alloys with respect to workability, strength and welding parameter optimization, Ain Shams Eng. J. 12 (1) (2021) 1143–1152, https://doi.org/10.1016/j.asej.2020.05.013. Q. Wang, et al., A tutorial on deep learning-based data analytics in manufacturing through a welding case study, J. Manuf. Process. 63 (2021) 2–13, https://doi. org/10.1016/j.jmapro.2020.04.044. J. Deepak, et al., Non-destructive testing (NDT) techniques for low carbon steel welded joints: a review and experimental study, Mater. Today: Proc. 44 (5) (2021) 3732–3737, https://doi.org/10.1016/j.matpr.2020.11.578. S.K. Dwivedi, M. Vishwakarma, A. Soni, Advances and researches on non destructive testing: a review, Mater. Today: Proc. 5 (2) (2018) 3690–3698, https://doi. org/10.1016/j.matpr.2017.11.620. M. Shaloo, et al., A review of non-destructive testing (NDT) techniques for defect detection: application to fusion welding and future Wire arc additive manufacturing processes, Materials 15 (10) (2022) 3697, https://doi.org/10.3390/ma15103697. S. Eckel, et al., Radiographic film system classification and noise characterisation by a camera-based digitisation procedure, NDT E Int. 111 (2020) 102241, https://doi.org/10.1016/j.ndteint.2020.102241. J. Szusta, et al., Effect of welding process parameters on the strength of dissimilar joints of S355 and Strenx 700 steels used in the Manufacture of Agricultural Machinery, Materials 16 (21) (2023) 6963, https://doi.org/10.3390/ma16216963. W. Hou, et al., Review on computer aided weld defect detection from radiography images, Appl. Sci. 10 (5) (2020) 1878, https://doi.org/10.3390/ app10051878. E. Yahaghi, M. Mirzapour, A. Movafeghi, Comparison of traditional and adaptive multi-scale products thresholding for enhancing the radiographs of welded object, Eur. Phys. J. Plus 136 (7) (2021) 744, https://doi.org/10.1140/epjp/s13360-021-01733-0. M. Prunella, et al., Deep learning for automatic vision-based recognition of industrial surface defects: a survey, IEEE Access 11 (2023) 43370–43423, https:// doi.org/10.1109/ACCESS.2023.3271748. W. Dai, et al., Deep learning assisted vision inspection of resistance spot welds, J. Manuf. Process. 62 (2021) 262–274, https://doi.org/10.1016/j. jmapro.2020.12.015. M.K. Ferguson, et al., Detection and segmentation of manufacturing defects with convolutional neural networks and transfer learning, Smart and sustainable manufacturing systems 2 (1) (2018) 20180033, https://doi.org/10.48550/arXiv.1808.02518. Y. Han, J. Fan, X. Yang, A structured light vision sensor for on-line weld bead measurement and weld quality inspection, Int. J. Adv. Des. Manuf. Technol. 106 (2020) 2065–2078, https://doi.org/10.1007/s00170-019-04450-2. D. Wang, et al., Deep network-assisted quality inspection of laser welding on power Battery, Sensors 23 (21) (2023) 8894, https://doi.org/10.3390/s23218894. D.P. Ramírez, et al., Pore segmentation in industrial radiographic images using adaptive thresholding and Morphological analysis, Trends in Agricultural and Environmental Sciences (2023). https://orcid.org/0000-0002-6485-9231. D.M. Hermosilla, et al., Shallow convolutional network excel for classifying motor imagery EEG in BCI applications, IEEE Access 9 (2021) 98275–98286, https://doi.org/10.1109/ACCESS.2021.3091399. D. Mery, et al., Pattern recognition in the automatic inspection of aluminium castings, Insight-Non-Destructive Testing and Condition Monitoring 45 (7) (2023) 475–483, https://doi.org/10.1784/insi.45.7.475.54452. T. W. Liao, D.-M. Li, Y.-M. Li, Detection of welding flaws from radiographic images with fuzzy clustering methods, Fuzzy Set Syst., 108 (2) (199) 145-158. https://doi.org/10.1016/S0165-0114(97)00307-2. T.W. Liao, Classification of welding flaw types with fuzzy expert systems, Expert Syst. Appl. 25 (1) (2003) 101–111, https://doi.org/10.1016/S0957-4174(03) 00010-1. R.R. da Silva, et al., Pattern recognition of weld defects detected by radiographic test, NDT E Int. 37 (6) (2004) 461–470, https://doi.org/10.1016/j. ndteint.2003.12.004. K. Carvajal, et al., Neural network method for failure detection with skewed class distribution, Insight-Non-Destructive Testing and Condition Monitoring 46 (7) (2004) 399–402, https://doi.org/10.1784/insi.46.7.399.55578. L. Yang, et al., Inspection of welding defect based on multi-feature fusion and a convolutional network, J. Nondestr. Eval. 40 (2021) 1–11, https://doi.org/ 10.1007/s10921-021-00823-4 S. Wang, Automatic detection and classification of steel surface defect using deep convolutional neural networks, Metals 11 (3) (2021) 388, https://doi.org/ 10.3390/met11030388 D. Say, et al., Automated categorization of multiclass welding defects using the x-ray image augmentation and convolutional neural network, Sensors 23 (14) (2023) 6422, https://doi.org/10.3390/s23146422. Ross Veitía, Deep Learning for Quality Prediction in Dissimilar Spot Welding DP600-Aisi304, Using a Convolutional Neural Network and Infrared Image Processing, 2020, https://doi.org/10.46354/i3m.2020.emss.057. R. Zhang, et al., Research on an ultrasonic detection method for weld defects based on neural network architecture search, Measurement 221 (2023) 113483, https://doi.org/10.1016/j.measurement.2023.113483. Z. Li, F. Liu, W. Yang, S. Peng, J. Zhou, A survey of convolutional neural networks: analysis, applications, and Prospects, IEEE Transact. Neural Networks Learn. Syst. 33 (12) (2022) 6999–7019, https://doi.org/10.1109/TNNLS.2021.3084827. K. Lee, et al., Review on the recent welding research with application of CNN-based deep learning part II: model evaluation and visualizations, Journal of Welding and Joining 39 (1) (2021) 20–26, https://doi.org/10.5781/JWJ.2021.39.1.2. R.V. Patil, Y.P. Reddy, Multiform weld joint flaws detection and classification by sagacious artificial neural network technique, Int. J. Adv. Des. Manuf. Technol. 125 (1) (2023) 913–943, https://doi.org/10.1007/s00170-022-10719-w. S. Kumaresan, et al., Deep learning based Simple CNN weld defects classification using optimization technique, Russ. J. Nondestr. Test. 58 (6) (2022) 499–509, https://doi.org/10.1134/S1061830922060109. A. Singh, K. Raj, T. Kumar, S. Verma, A.M. Roy, Deep learning-based Cost-effective and Responsive Robot for autism Treatment, Drones 7 (2) (2023) 81, https:// doi.org/10.3390/drones7020081. S. Perri, et al., Welding defects classification through a convolutional neural network, Manufacturing Letters 35 (2023) 29–32, https://doi.org/10.1016/j. mfglet.2022.11.006. D. Mery, et al., The database of X-ray images for nondestructive testing, J. Nondestr. Eval. 34 (4) (2015) 42, https://doi.org/10.1007/s10921-015-0315-7. W. Guo, H. Qu, L. Liang, WDXI: the dataset of X-ray image for weld defects, in: 2018 14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD), 2018, pp. 1051–1055, https://doi.org/10.1109/FSKD.2018.8686975. B. Totino, F. Spagnolo, S. Perri, RIAWELC: a Novel dataset of radiographic images for automatic weld defects classification, International Journal of Electrical and Computer Engineering Research 3 (1) (2023) 13–17, https://doi.org/10.53375/ijecer.2023.320. S. Kumaresan, et al., Transfer learning with CNN for classification of weld defect, IEEE Access 9 (2021) 95097–95108, https://doi.org/10.1109/ ACCESS.2021.3093487. H.G. Hernandez-Palma, et al., Technological tools based on artificial intelligence in the sugar industry: a Bibliometric analysis and future Perspectives for energy efficiency, LADEE 4 (2) (2023) 49–64, https://doi.org/10.17981/ladee.04.02.2023.4. S. Kumaresan, et al., Deep learning-based weld defect classification using VGG16 transfer learning adaptive fine-tuning, Int. J. Interact. Des. Manuf. 17 (6) (2023) 2999–3010, https://doi.org/10.1007/s12008-023-01327-3. M. Hussain, J.J. Bird, D.R. Faria, A study on CNN transfer learning for image classification, Advances in Computational Intelligence Systems, Adv. Intell. Syst. Comput. 840 (2019), https://doi.org/10.1007/978-3-319-97982-3_16. Springer, Cham. W. Jiao, et al., End-to-end prediction of weld penetration: a deep learning and transfer learning based method, J. Manuf. Process. 63 (2021) 191–197, https:// doi.org/10.1016/j.jmapro.2020.01.044. D.D. Kumar, et al., Semi-supervised transfer learning-based automatic weld defect detection and visual inspection, Eng. Struct. 292 (2023) 116580, https://doi. org/10.1016/j.engstruct.2023.116580. H. Pan, et al., A new image recognition and classification method combining transfer learning algorithm and mobilenet model for welding defects, IEEE Access 8 (2020) 119951–119960, https://doi.org/10.1109/ACCESS.2020.3005450. A.M. Roy, R. Boseand J. Bhaduri, A fast accurate fine-grain object detection model based on YOLOv4 deep neural network, Neural Comput. Appl. 34 (5) (2022) 3895–3921, https://doi.org/10.1007/s00521-021-06651-x. A.M. Roy, J. Bhaduri, DenseSPH-YOLOv5: an automated damage detection model based on DenseNet and Swin-Transformer prediction head-enabled YOLOv5 with attention mechanism, Adv. Eng. Inf. 56 (2023) 102007, https://doi.org/10.1016/j.aei.2023.102007. D.V. Rayudu, J.F. Roseline, Accurate Weather Forecasting for Rainfall prediction using artificial neural network compared with deep learning neural network, in: 2023 International Conference on Artificial Intelligence and Knowledge Discovery in Concurrent Engineering (ICECONF), 2023, pp. 1–6, https://doi.org/ 10.1109/ICECONF57129.2023.10084252. Chennai, India. X. Wang, X. Yu, Understanding the effect of transfer learning on the automatic welding defect detection, NDT E Int. 134 (2023) 102784, https://doi.org/ 10.1016/j.ndteint.2022.102784. J. Nunez, ˜ et al., Design of a fuzzy controller for a hybrid generation system, IOP Conf. Ser. Mater. Sci. Eng. 844 (2020) 012017, https://doi.org/10.1088/1757- 899X/844/1/012017. B. Jiang, S. Chen, B. Wang, B. Luo, MGLNN: Semi-supervised learning via Multiple Graph Cooperative learning neural networks, Neural Network. 153 (2022) 204–214, https://doi.org/10.1016/j.neunet.2022.05.024. Z. Li, Y. Li, Y. Liu, P. Wang, R. Lu, H.B. Gooi, Deep learning based densely connected network for load Forecasting, IEEE Trans. Power Syst. 36 (4) (2021) 2829–2840, https://doi.org/10.1109/TPWRS.2020.3048359. A. Singh, L. Bruzzone, Mono- and Dual-Regulated Contractive-Expansive-Contractive deep convolutional networks for classification of Multispectral Remote sensing images, Geosci. Rem. Sens. Lett. IEEE 19 (2022) 5513605, https://doi.org/10.1109/LGRS.2022.3211861, 1-5. B. Liu, et al., Weld Defect Images Classification with Vgg16-Based Neural Network, International Forum on Digital TV and Wireless Multimedia Communications, vol. 815, Springer, Singapore, 2018, https://doi.org/10.1007/978-981-10-8108-8_20. L. Mohanasundari, Performance analysis of weld image classification using modified Resnet CNN architecture, Turkish Journal of Computer and Mathematics Education (TURCOMAT) 12 (2) (2021) 2260–2266. https://turcomat.org/index.php/turkbilmat/article/view/1943. V.A. Golodov, A.A. Mittseva, Weld segmentation and defect detection in radiographic images of Pipe welds, in: 2019 International Russian Automation Conference (RusAutoCon), Sochi, Russia, 2019, pp. 1–6, https://doi.org/10.1109/RUSAUTOCON.2019.8867734. J. Deng, W. Dong, R. Socher, L.-J. Li, Kai Li, Li Fei-Fei, ImageNet: a large-scale hierarchical image database, in: 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009, pp. 248–255, https://doi.org/10.1109/CVPR.2009.5206848. Miami, FL, USA. D. Chauveau, Review of NDT and process monitoring techniques useable to produce high-quality parts by welding or additive manufacturing, Weld. World 62 (5) (2018) 1097–1118, https://doi.org/10.1007/s40194-018-0609-3. J. Rao, et al., Non-destructive testing of metal-based additively manufactured parts and processes: a review, Virtual Phys. Prototyp. 18 (1) (2023) e2266658, https://doi.org/10.1080/17452759.2023.2266658. D. Priyasudana, et al., Double side friction stir welding effect on mechanical properties and corrosion rate of aluminum alloy AA6061, Heliyon 9 (2) (2023) e13366, https://doi.org/10.1016/j.heliyon.2023.e13366 M.N. Ilman, Microstructure and mechanical properties of friction stir spot welded AA5052-H112 aluminum alloy, Heliyon 7 (2) (2021) e06009, https://doi.org/ 10.1016/j.heliyon.2021.e06009. S. Shin, C. Jin, J. Yuand S. Rhee, Real-time detection of weld defects for automated welding process base on deep neural network, Metals 10 (3) (2020) 389–2020, https://doi.org/10.3390/met10030389. D.-Y. Kim, et al., Weld fatigue behavior of gas metal arc welded steel sheets based on porosity and gap size, Int. J. Adv. Des. Manuf. Technol. 124 (3) (2023) 1141–1153, https://doi.org/10.1007/s00170-022-10567-8. T. Fujii, N. Ogasawara, K. Tohgo, Y. Shimamura, Monte Carlo simulation of stress corrosion cracking in welded metal with surface defects and life estimation, Int. J. Mech. Sci. 270 (2024) 109079, https://doi.org/10.1016/j.ijmecsci.2024.109079. Y. Li, et al., Detection model of invisible weld defects using magneto-optical imaging induced by rotating magnetic field, in: 2019 IEEE 15th International Conference on Automation Science and Engineering (CASE), 2019, pp. 1–5, https://doi.org/10.1109/COASE.2019.8843280. Vancouver, BC, Canada. H.M. Rai, K. Chatterjee, Hybrid CNN-LSTM deep learning model and ensemble technique for automatic detection of myocardial infarction using big ECG data, Appl. Intell. 52 (5) (2022) 5366–5384, https://doi.org/10.1007/s10489-021-02696-6. Z. Zhu, K. Lin, A.K. Jain, J. Zhou, Transfer learning in deep Reinforcement learning: a survey, IEEE Trans. Pattern Anal. Mach. Intell. 45 11 (2023) 13344–13362, https://doi.org/10.1109/TPAMI.2023.3292075. W. Wang, G. Wen, Z. Zheng, Design of deep learning Mixed Language short Text Sentiment classification system based on CNN algorithm, in: 2022 IEEE 2nd International Conference on Mobile Networks and Wireless Communications (ICMNWC), Tumkur, 2022, pp. 1–5, https://doi.org/10.1109/ ICMNWC56175.2022.10031786. Karnataka, India. K. Pal, B.V. Patel, Data classification with K-fold cross validation and Holdout accuracy estimation methods with 5 different machine learning techniques, in: 2020 Fourth International Conference on Computing Methodologies and Communication (ICCMC), 2020, pp. 83–87, https://doi.org/10.1109/ ICCMC48092.2020.ICCMC-00016. Erode, India. S. Szeghalmy, A. Fazekas, A comparative study of the use of stratified cross-validation and distribution-balanced stratified cross-validation in imbalanced learning, Sensors 23 (4) (2023) 2333, https://doi.org/10.3390/s23042333. J.-R. Lee, K.-W. Ng, Y.-J. Yoong, Face and facial expressions recognition system for blind people using ResNet50 architecture and CNN, Journal of Informatics and Web Engineering 2 (2) (2023) 284–298, https://doi.org/10.33093/jiwe.2023.2.2.20. S. Mascarenhas, M. Agarwal, A comparison between VGG16, VGG19 and ResNet50 architecture frameworks for Image Classification, in: 2021 International Conference on Disruptive Technologies for Multi-Disciplinary Research and Applications (CENTCON), 2021, pp. 96–99, https://doi.org/10.1109/ CENTCON52345.2021.9687944. Bengaluru, India. S. Tofigh, M.O. Ahmad, M.N.S. Swamy, A low-Complexity modified ThiNet algorithm for Pruning convolutional neural networks, IEEE Signal Process. Lett. 29 (2022) 1012–1016, https://doi.org/10.1109/LSP.2022.3164328. X. Huang, High resolution Remote sensing image classification based on deep transfer learning and multi feature network, IEEE Access 11 (2023) 110075–110085, https://doi.org/10.1109/ACCESS.2023.3320792 D. Milan´es-Hermosilla, et al., Monte Carlo dropout for uncertainty estimation and motor imagery classification, Sensors 21 (21) (2021) 7241, https://doi.org/ 10.3390/s21217241. P. Dileep, D. Das, P.K. Bora, Dense layer dropout based CNN architecture for automatic Modulation classification, in: 2020 National Conference on Communications (NCC), Kharagpur, India, 2020, pp. 1–5, https://doi.org/10.1109/NCC48643.2020.9055989. Y. Liu, et al., Guided dropout: Improving deep networks without increased computation, Intelligent Automation & Soft Computing 36 (3) (2023) 2519–2528, https://doi.org/10.32604/iasc.2023.033286. |
| dc.relation.citationendpage.none.fl_str_mv |
11 |
| dc.relation.citationstartpage.none.fl_str_mv |
1 |
| dc.relation.citationissue.none.fl_str_mv |
e30590 |
| dc.relation.citationvolume.none.fl_str_mv |
10 |
| dc.rights.none.fl_str_mv |
© 2024 The Authors. |
| dc.rights.license.none.fl_str_mv |
Atribución 4.0 Internacional (CC BY 4.0) |
| dc.rights.uri.none.fl_str_mv |
https://creativecommons.org/licenses/by/4.0/ |
| dc.rights.accessrights.none.fl_str_mv |
info:eu-repo/semantics/openAccess |
| dc.rights.coar.none.fl_str_mv |
http://purl.org/coar/access_right/c_abf2 |
| rights_invalid_str_mv |
Atribución 4.0 Internacional (CC BY 4.0) © 2024 The Authors. https://creativecommons.org/licenses/by/4.0/ http://purl.org/coar/access_right/c_abf2 |
| eu_rights_str_mv |
openAccess |
| dc.format.extent.none.fl_str_mv |
11 páginas |
| dc.format.mimetype.none.fl_str_mv |
application/pdf |
| dc.publisher.none.fl_str_mv |
Elsevier B.V. |
| dc.publisher.place.none.fl_str_mv |
Netherlands |
| publisher.none.fl_str_mv |
Elsevier B.V. |
| dc.source.none.fl_str_mv |
https://www.sciencedirect.com/science/article/pii/S2405844024066210?via%3Dihub |
| institution |
Corporación Universidad de la Costa |
| bitstream.url.fl_str_mv |
https://repositorio.cuc.edu.co/bitstreams/73034bd9-e6e6-47a9-a77d-e8ddf56c269e/download https://repositorio.cuc.edu.co/bitstreams/4d4cafbd-29e8-47f2-a9b4-2c65950eb255/download https://repositorio.cuc.edu.co/bitstreams/6c08a334-e79b-4e01-9c1f-bedee7e872b9/download https://repositorio.cuc.edu.co/bitstreams/e2259fcb-344b-4cd3-82dc-c48f399334ad/download https://repositorio.cuc.edu.co/bitstreams/46adc435-cc66-4660-8cd1-3ff0d378304b/download https://repositorio.cuc.edu.co/bitstreams/96b17128-4a79-4d53-82cb-336366de287b/download |
| bitstream.checksum.fl_str_mv |
73a5432e0b76442b22b026844140d683 ea07c8decdd3c0750a347784f608a5e9 6445470caa422650a497736ee2eb5dd4 c53b5c49fb8018e8b56c4f9d68e0733f ca3a946794984c14acec968f7d0fec6f c62e9c2237da5cbae6e9ee1f48896c78 |
| bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 MD5 |
| repository.name.fl_str_mv |
Repositorio de la Universidad de la Costa CUC |
| repository.mail.fl_str_mv |
repdigital@cuc.edu.co |
| _version_ |
1828166536403091456 |
| spelling |
Atribución 4.0 Internacional (CC BY 4.0)© 2024 The Authors.https://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccesshttp://purl.org/coar/access_right/c_abf2Palma Ramírez, DayanaRoss Veitía, Bárbara D.Font Ariosa, PabloEspinel Hernández, AlejandroSánchez Roca, ÁngelCarvajar Fals, HipólitoNuñez Álvarez, José R.Hernández Herrera, Hernan2024-10-23T12:40:23Z2024-10-23T12:40:23Z2024-05-15Dayana Palma-Ramírez, Bárbara D. Ross-Veitía, Pablo Font-Ariosa, Alejandro Espinel-Hernández, Angel Sanchez-Roca, Hipólito Carvajal-Fals, José R. Nuñez-Alvarez, Hernan Hernández-Herrera, Deep convolutional neural network for weld defect classification in radiographic images, Heliyon, Volume 10, Issue 9, 2024, e30590, ISSN 2405-8440, https://doi.org/10.1016/j.heliyon.2024.e30590.https://hdl.handle.net/11323/1348110.1016/j.heliyon.2024.e305902405-8440Corporación Universidad de la CostaREDICUC - Repositorio CUChttps://repositorio.cuc.edu.co/The quality of welds is critical to the safety of structures in construction, so early detection of irregularities is crucial. Advances in machine vision inspection technologies, such as deep learning models, have improved the detection of weld defects. This paper presents a new CNN model based on ResNet50 to classify four types of weld defects in radiographic images: crack, pore, non-penetration, and no defect. Stratified cross-validation, data augmentation, and regularization were used to improve generalization and avoid over-fitting. The model was tested on three datasets, RIAWELC, GDXray, and a private dataset of low image quality, obtaining an accuracy of 98.75 %, 90.255 %, and 75.83 %, respectively. The model proposed in this paper achieves high accuracies on different datasets and constitutes a valuable tool to improve the efficiency and effectiveness of quality control processes in the welding industry. Moreover, experimental tests show that the proposed approach performs well on even low-resolution images.11 páginasapplication/pdfengElsevier B.V.Netherlandshttps://www.sciencedirect.com/science/article/pii/S2405844024066210?via%3DihubDeep convolutional neural network for weld defect classification in radiographic imagesArtículo de revistahttp://purl.org/coar/resource_type/c_2df8fbb1Textinfo:eu-repo/semantics/articlehttp://purl.org/redcol/resource_type/ARTinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/version/c_970fb48d4fbd8a85HeliyonT.J. McPheron, R.M. Stwalley, Engineering Challenges Associated with Welding Field Repairs, Engineering Principles-Welding and Residual Stresses, 2022, https://doi.org/10.5772/intechopen.104263.M. P´erez de la Parte, et al., Effect of zinc coating on delay nugget formation in dissimilar DP600-AISI304 welded joints obtained by the resistance spot welding process, Int. J. Adv. Des. Manuf. Technol. 120 (2022) 1877–1887, https://doi.org/10.1007/s00170-022-08849-2.D. Varshney, K. Kumar, Application and use of different aluminium alloys with respect to workability, strength and welding parameter optimization, Ain Shams Eng. J. 12 (1) (2021) 1143–1152, https://doi.org/10.1016/j.asej.2020.05.013.Q. Wang, et al., A tutorial on deep learning-based data analytics in manufacturing through a welding case study, J. Manuf. Process. 63 (2021) 2–13, https://doi. org/10.1016/j.jmapro.2020.04.044.J. Deepak, et al., Non-destructive testing (NDT) techniques for low carbon steel welded joints: a review and experimental study, Mater. Today: Proc. 44 (5) (2021) 3732–3737, https://doi.org/10.1016/j.matpr.2020.11.578.S.K. Dwivedi, M. Vishwakarma, A. Soni, Advances and researches on non destructive testing: a review, Mater. Today: Proc. 5 (2) (2018) 3690–3698, https://doi. org/10.1016/j.matpr.2017.11.620.M. Shaloo, et al., A review of non-destructive testing (NDT) techniques for defect detection: application to fusion welding and future Wire arc additive manufacturing processes, Materials 15 (10) (2022) 3697, https://doi.org/10.3390/ma15103697.S. Eckel, et al., Radiographic film system classification and noise characterisation by a camera-based digitisation procedure, NDT E Int. 111 (2020) 102241, https://doi.org/10.1016/j.ndteint.2020.102241.J. Szusta, et al., Effect of welding process parameters on the strength of dissimilar joints of S355 and Strenx 700 steels used in the Manufacture of Agricultural Machinery, Materials 16 (21) (2023) 6963, https://doi.org/10.3390/ma16216963.W. Hou, et al., Review on computer aided weld defect detection from radiography images, Appl. Sci. 10 (5) (2020) 1878, https://doi.org/10.3390/ app10051878.E. Yahaghi, M. Mirzapour, A. Movafeghi, Comparison of traditional and adaptive multi-scale products thresholding for enhancing the radiographs of welded object, Eur. Phys. J. Plus 136 (7) (2021) 744, https://doi.org/10.1140/epjp/s13360-021-01733-0.M. Prunella, et al., Deep learning for automatic vision-based recognition of industrial surface defects: a survey, IEEE Access 11 (2023) 43370–43423, https:// doi.org/10.1109/ACCESS.2023.3271748.W. Dai, et al., Deep learning assisted vision inspection of resistance spot welds, J. Manuf. Process. 62 (2021) 262–274, https://doi.org/10.1016/j. jmapro.2020.12.015.M.K. Ferguson, et al., Detection and segmentation of manufacturing defects with convolutional neural networks and transfer learning, Smart and sustainable manufacturing systems 2 (1) (2018) 20180033, https://doi.org/10.48550/arXiv.1808.02518.Y. Han, J. Fan, X. Yang, A structured light vision sensor for on-line weld bead measurement and weld quality inspection, Int. J. Adv. Des. Manuf. Technol. 106 (2020) 2065–2078, https://doi.org/10.1007/s00170-019-04450-2.D. Wang, et al., Deep network-assisted quality inspection of laser welding on power Battery, Sensors 23 (21) (2023) 8894, https://doi.org/10.3390/s23218894.D.P. Ramírez, et al., Pore segmentation in industrial radiographic images using adaptive thresholding and Morphological analysis, Trends in Agricultural and Environmental Sciences (2023). https://orcid.org/0000-0002-6485-9231.D.M. Hermosilla, et al., Shallow convolutional network excel for classifying motor imagery EEG in BCI applications, IEEE Access 9 (2021) 98275–98286, https://doi.org/10.1109/ACCESS.2021.3091399.D. Mery, et al., Pattern recognition in the automatic inspection of aluminium castings, Insight-Non-Destructive Testing and Condition Monitoring 45 (7) (2023) 475–483, https://doi.org/10.1784/insi.45.7.475.54452.T. W. Liao, D.-M. Li, Y.-M. Li, Detection of welding flaws from radiographic images with fuzzy clustering methods, Fuzzy Set Syst., 108 (2) (199) 145-158. https://doi.org/10.1016/S0165-0114(97)00307-2.T.W. Liao, Classification of welding flaw types with fuzzy expert systems, Expert Syst. Appl. 25 (1) (2003) 101–111, https://doi.org/10.1016/S0957-4174(03) 00010-1.R.R. da Silva, et al., Pattern recognition of weld defects detected by radiographic test, NDT E Int. 37 (6) (2004) 461–470, https://doi.org/10.1016/j. ndteint.2003.12.004.K. Carvajal, et al., Neural network method for failure detection with skewed class distribution, Insight-Non-Destructive Testing and Condition Monitoring 46 (7) (2004) 399–402, https://doi.org/10.1784/insi.46.7.399.55578.L. Yang, et al., Inspection of welding defect based on multi-feature fusion and a convolutional network, J. Nondestr. Eval. 40 (2021) 1–11, https://doi.org/ 10.1007/s10921-021-00823-4S. Wang, Automatic detection and classification of steel surface defect using deep convolutional neural networks, Metals 11 (3) (2021) 388, https://doi.org/ 10.3390/met11030388D. Say, et al., Automated categorization of multiclass welding defects using the x-ray image augmentation and convolutional neural network, Sensors 23 (14) (2023) 6422, https://doi.org/10.3390/s23146422.Ross Veitía, Deep Learning for Quality Prediction in Dissimilar Spot Welding DP600-Aisi304, Using a Convolutional Neural Network and Infrared Image Processing, 2020, https://doi.org/10.46354/i3m.2020.emss.057.R. Zhang, et al., Research on an ultrasonic detection method for weld defects based on neural network architecture search, Measurement 221 (2023) 113483, https://doi.org/10.1016/j.measurement.2023.113483.Z. Li, F. Liu, W. Yang, S. Peng, J. Zhou, A survey of convolutional neural networks: analysis, applications, and Prospects, IEEE Transact. Neural Networks Learn. Syst. 33 (12) (2022) 6999–7019, https://doi.org/10.1109/TNNLS.2021.3084827.K. Lee, et al., Review on the recent welding research with application of CNN-based deep learning part II: model evaluation and visualizations, Journal of Welding and Joining 39 (1) (2021) 20–26, https://doi.org/10.5781/JWJ.2021.39.1.2.R.V. Patil, Y.P. Reddy, Multiform weld joint flaws detection and classification by sagacious artificial neural network technique, Int. J. Adv. Des. Manuf. Technol. 125 (1) (2023) 913–943, https://doi.org/10.1007/s00170-022-10719-w.S. Kumaresan, et al., Deep learning based Simple CNN weld defects classification using optimization technique, Russ. J. Nondestr. Test. 58 (6) (2022) 499–509, https://doi.org/10.1134/S1061830922060109.A. Singh, K. Raj, T. Kumar, S. Verma, A.M. Roy, Deep learning-based Cost-effective and Responsive Robot for autism Treatment, Drones 7 (2) (2023) 81, https:// doi.org/10.3390/drones7020081.S. Perri, et al., Welding defects classification through a convolutional neural network, Manufacturing Letters 35 (2023) 29–32, https://doi.org/10.1016/j. mfglet.2022.11.006.D. Mery, et al., The database of X-ray images for nondestructive testing, J. Nondestr. Eval. 34 (4) (2015) 42, https://doi.org/10.1007/s10921-015-0315-7.W. Guo, H. Qu, L. Liang, WDXI: the dataset of X-ray image for weld defects, in: 2018 14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD), 2018, pp. 1051–1055, https://doi.org/10.1109/FSKD.2018.8686975.B. Totino, F. Spagnolo, S. Perri, RIAWELC: a Novel dataset of radiographic images for automatic weld defects classification, International Journal of Electrical and Computer Engineering Research 3 (1) (2023) 13–17, https://doi.org/10.53375/ijecer.2023.320.S. Kumaresan, et al., Transfer learning with CNN for classification of weld defect, IEEE Access 9 (2021) 95097–95108, https://doi.org/10.1109/ ACCESS.2021.3093487.H.G. Hernandez-Palma, et al., Technological tools based on artificial intelligence in the sugar industry: a Bibliometric analysis and future Perspectives for energy efficiency, LADEE 4 (2) (2023) 49–64, https://doi.org/10.17981/ladee.04.02.2023.4.S. Kumaresan, et al., Deep learning-based weld defect classification using VGG16 transfer learning adaptive fine-tuning, Int. J. Interact. Des. Manuf. 17 (6) (2023) 2999–3010, https://doi.org/10.1007/s12008-023-01327-3.M. Hussain, J.J. Bird, D.R. Faria, A study on CNN transfer learning for image classification, Advances in Computational Intelligence Systems, Adv. Intell. Syst. Comput. 840 (2019), https://doi.org/10.1007/978-3-319-97982-3_16. Springer, Cham.W. Jiao, et al., End-to-end prediction of weld penetration: a deep learning and transfer learning based method, J. Manuf. Process. 63 (2021) 191–197, https:// doi.org/10.1016/j.jmapro.2020.01.044.D.D. Kumar, et al., Semi-supervised transfer learning-based automatic weld defect detection and visual inspection, Eng. Struct. 292 (2023) 116580, https://doi. org/10.1016/j.engstruct.2023.116580.H. Pan, et al., A new image recognition and classification method combining transfer learning algorithm and mobilenet model for welding defects, IEEE Access 8 (2020) 119951–119960, https://doi.org/10.1109/ACCESS.2020.3005450.A.M. Roy, R. Boseand J. Bhaduri, A fast accurate fine-grain object detection model based on YOLOv4 deep neural network, Neural Comput. Appl. 34 (5) (2022) 3895–3921, https://doi.org/10.1007/s00521-021-06651-x.A.M. Roy, J. Bhaduri, DenseSPH-YOLOv5: an automated damage detection model based on DenseNet and Swin-Transformer prediction head-enabled YOLOv5 with attention mechanism, Adv. Eng. Inf. 56 (2023) 102007, https://doi.org/10.1016/j.aei.2023.102007.D.V. Rayudu, J.F. Roseline, Accurate Weather Forecasting for Rainfall prediction using artificial neural network compared with deep learning neural network, in: 2023 International Conference on Artificial Intelligence and Knowledge Discovery in Concurrent Engineering (ICECONF), 2023, pp. 1–6, https://doi.org/ 10.1109/ICECONF57129.2023.10084252. Chennai, India.X. Wang, X. Yu, Understanding the effect of transfer learning on the automatic welding defect detection, NDT E Int. 134 (2023) 102784, https://doi.org/ 10.1016/j.ndteint.2022.102784.J. Nunez, ˜ et al., Design of a fuzzy controller for a hybrid generation system, IOP Conf. Ser. Mater. Sci. Eng. 844 (2020) 012017, https://doi.org/10.1088/1757- 899X/844/1/012017.B. Jiang, S. Chen, B. Wang, B. Luo, MGLNN: Semi-supervised learning via Multiple Graph Cooperative learning neural networks, Neural Network. 153 (2022) 204–214, https://doi.org/10.1016/j.neunet.2022.05.024.Z. Li, Y. Li, Y. Liu, P. Wang, R. Lu, H.B. Gooi, Deep learning based densely connected network for load Forecasting, IEEE Trans. Power Syst. 36 (4) (2021) 2829–2840, https://doi.org/10.1109/TPWRS.2020.3048359.A. Singh, L. Bruzzone, Mono- and Dual-Regulated Contractive-Expansive-Contractive deep convolutional networks for classification of Multispectral Remote sensing images, Geosci. Rem. Sens. Lett. IEEE 19 (2022) 5513605, https://doi.org/10.1109/LGRS.2022.3211861, 1-5.B. Liu, et al., Weld Defect Images Classification with Vgg16-Based Neural Network, International Forum on Digital TV and Wireless Multimedia Communications, vol. 815, Springer, Singapore, 2018, https://doi.org/10.1007/978-981-10-8108-8_20.L. Mohanasundari, Performance analysis of weld image classification using modified Resnet CNN architecture, Turkish Journal of Computer and Mathematics Education (TURCOMAT) 12 (2) (2021) 2260–2266. https://turcomat.org/index.php/turkbilmat/article/view/1943.V.A. Golodov, A.A. Mittseva, Weld segmentation and defect detection in radiographic images of Pipe welds, in: 2019 International Russian Automation Conference (RusAutoCon), Sochi, Russia, 2019, pp. 1–6, https://doi.org/10.1109/RUSAUTOCON.2019.8867734.J. Deng, W. Dong, R. Socher, L.-J. Li, Kai Li, Li Fei-Fei, ImageNet: a large-scale hierarchical image database, in: 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009, pp. 248–255, https://doi.org/10.1109/CVPR.2009.5206848. Miami, FL, USA.D. Chauveau, Review of NDT and process monitoring techniques useable to produce high-quality parts by welding or additive manufacturing, Weld. World 62 (5) (2018) 1097–1118, https://doi.org/10.1007/s40194-018-0609-3.J. Rao, et al., Non-destructive testing of metal-based additively manufactured parts and processes: a review, Virtual Phys. Prototyp. 18 (1) (2023) e2266658, https://doi.org/10.1080/17452759.2023.2266658.D. Priyasudana, et al., Double side friction stir welding effect on mechanical properties and corrosion rate of aluminum alloy AA6061, Heliyon 9 (2) (2023) e13366, https://doi.org/10.1016/j.heliyon.2023.e13366M.N. Ilman, Microstructure and mechanical properties of friction stir spot welded AA5052-H112 aluminum alloy, Heliyon 7 (2) (2021) e06009, https://doi.org/ 10.1016/j.heliyon.2021.e06009.S. Shin, C. Jin, J. Yuand S. Rhee, Real-time detection of weld defects for automated welding process base on deep neural network, Metals 10 (3) (2020) 389–2020, https://doi.org/10.3390/met10030389.D.-Y. Kim, et al., Weld fatigue behavior of gas metal arc welded steel sheets based on porosity and gap size, Int. J. Adv. Des. Manuf. Technol. 124 (3) (2023) 1141–1153, https://doi.org/10.1007/s00170-022-10567-8.T. Fujii, N. Ogasawara, K. Tohgo, Y. Shimamura, Monte Carlo simulation of stress corrosion cracking in welded metal with surface defects and life estimation, Int. J. Mech. Sci. 270 (2024) 109079, https://doi.org/10.1016/j.ijmecsci.2024.109079.Y. Li, et al., Detection model of invisible weld defects using magneto-optical imaging induced by rotating magnetic field, in: 2019 IEEE 15th International Conference on Automation Science and Engineering (CASE), 2019, pp. 1–5, https://doi.org/10.1109/COASE.2019.8843280. Vancouver, BC, Canada.H.M. Rai, K. Chatterjee, Hybrid CNN-LSTM deep learning model and ensemble technique for automatic detection of myocardial infarction using big ECG data, Appl. Intell. 52 (5) (2022) 5366–5384, https://doi.org/10.1007/s10489-021-02696-6.Z. Zhu, K. Lin, A.K. Jain, J. Zhou, Transfer learning in deep Reinforcement learning: a survey, IEEE Trans. Pattern Anal. Mach. Intell. 45 11 (2023) 13344–13362, https://doi.org/10.1109/TPAMI.2023.3292075.W. Wang, G. Wen, Z. Zheng, Design of deep learning Mixed Language short Text Sentiment classification system based on CNN algorithm, in: 2022 IEEE 2nd International Conference on Mobile Networks and Wireless Communications (ICMNWC), Tumkur, 2022, pp. 1–5, https://doi.org/10.1109/ ICMNWC56175.2022.10031786. Karnataka, India.K. Pal, B.V. Patel, Data classification with K-fold cross validation and Holdout accuracy estimation methods with 5 different machine learning techniques, in: 2020 Fourth International Conference on Computing Methodologies and Communication (ICCMC), 2020, pp. 83–87, https://doi.org/10.1109/ ICCMC48092.2020.ICCMC-00016. Erode, India.S. Szeghalmy, A. Fazekas, A comparative study of the use of stratified cross-validation and distribution-balanced stratified cross-validation in imbalanced learning, Sensors 23 (4) (2023) 2333, https://doi.org/10.3390/s23042333.J.-R. Lee, K.-W. Ng, Y.-J. Yoong, Face and facial expressions recognition system for blind people using ResNet50 architecture and CNN, Journal of Informatics and Web Engineering 2 (2) (2023) 284–298, https://doi.org/10.33093/jiwe.2023.2.2.20.S. Mascarenhas, M. Agarwal, A comparison between VGG16, VGG19 and ResNet50 architecture frameworks for Image Classification, in: 2021 International Conference on Disruptive Technologies for Multi-Disciplinary Research and Applications (CENTCON), 2021, pp. 96–99, https://doi.org/10.1109/ CENTCON52345.2021.9687944. Bengaluru, India.S. Tofigh, M.O. Ahmad, M.N.S. Swamy, A low-Complexity modified ThiNet algorithm for Pruning convolutional neural networks, IEEE Signal Process. Lett. 29 (2022) 1012–1016, https://doi.org/10.1109/LSP.2022.3164328.X. Huang, High resolution Remote sensing image classification based on deep transfer learning and multi feature network, IEEE Access 11 (2023) 110075–110085, https://doi.org/10.1109/ACCESS.2023.3320792D. Milan´es-Hermosilla, et al., Monte Carlo dropout for uncertainty estimation and motor imagery classification, Sensors 21 (21) (2021) 7241, https://doi.org/ 10.3390/s21217241.P. Dileep, D. Das, P.K. Bora, Dense layer dropout based CNN architecture for automatic Modulation classification, in: 2020 National Conference on Communications (NCC), Kharagpur, India, 2020, pp. 1–5, https://doi.org/10.1109/NCC48643.2020.9055989.Y. Liu, et al., Guided dropout: Improving deep networks without increased computation, Intelligent Automation & Soft Computing 36 (3) (2023) 2519–2528, https://doi.org/10.32604/iasc.2023.033286.111e3059010Radiographic testingClassificationWeld defectsCNNsTransfer learningPublicationLICENSElicense.txtlicense.txttext/plain; charset=utf-815543https://repositorio.cuc.edu.co/bitstreams/73034bd9-e6e6-47a9-a77d-e8ddf56c269e/download73a5432e0b76442b22b026844140d683MD51ORIGINALDeep convolutional neural network for weld defect classification.htmDeep convolutional neural network for weld defect classification.htmtext/html187976https://repositorio.cuc.edu.co/bitstreams/4d4cafbd-29e8-47f2-a9b4-2c65950eb255/downloadea07c8decdd3c0750a347784f608a5e9MD52Deep convolutional neural network for weld defect classification in radiographic images.pdfDeep convolutional neural network for weld defect classification in radiographic images.pdfapplication/pdf2716365https://repositorio.cuc.edu.co/bitstreams/6c08a334-e79b-4e01-9c1f-bedee7e872b9/download6445470caa422650a497736ee2eb5dd4MD53TEXTDeep convolutional neural network for weld defect classification.htm.txtDeep convolutional neural network for weld defect classification.htm.txtExtracted texttext/plain2740https://repositorio.cuc.edu.co/bitstreams/e2259fcb-344b-4cd3-82dc-c48f399334ad/downloadc53b5c49fb8018e8b56c4f9d68e0733fMD54Deep convolutional neural network for weld defect classification in radiographic images.pdf.txtDeep convolutional neural network for weld defect classification in radiographic images.pdf.txtExtracted texttext/plain57286https://repositorio.cuc.edu.co/bitstreams/46adc435-cc66-4660-8cd1-3ff0d378304b/downloadca3a946794984c14acec968f7d0fec6fMD55THUMBNAILDeep convolutional neural network for weld defect classification in radiographic images.pdf.jpgDeep convolutional neural network for weld defect classification in radiographic images.pdf.jpgGenerated Thumbnailimage/jpeg13420https://repositorio.cuc.edu.co/bitstreams/96b17128-4a79-4d53-82cb-336366de287b/downloadc62e9c2237da5cbae6e9ee1f48896c78MD5611323/13481oai:repositorio.cuc.edu.co:11323/134812024-10-24 03:00:18.746https://creativecommons.org/licenses/by/4.0/© 2024 The Authors.open.accesshttps://repositorio.cuc.edu.coRepositorio de la Universidad de la Costa CUCrepdigital@cuc.edu.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 |