Deep convolutional neural network for weld defect classification in radiographic images

Palma-Ramírez, Dayana; Ross-Veitía, Bárbara D.; Font-Ariosa, Pablo; Espinel-Hernández, Alejandro; Sanchez-Roca, Angel; Carvajal-Fals, Hipólito; Nuñez-Alvarez, José R.; Hernández-Herrera, Hernan

doi:10.1016/j.heliyon.2024.e30590

Cited by 4 publications

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

A study on CT detection image generation based on decompound synthesize method

Fu,

Liu,

Zeng

et al. 2024

Journal of X-Ray Science and Technology: Clinical Applications

View full text Add to dashboard Cite

Background Nuclear graphite and carbon components are vital structural elements in the cores of high-temperature gas-cooled reactors(HTGR), serving crucial roles in neutron reflection, moderation, and insulation. The structural integrity and stable operation of these reactors heavily depend on the quality of these components. Helical Computed Tomography (CT) technology provides a method for detecting and intelligently identifying defects within these structures. However, the scarcity of defect datasets limits the performance of deep learning-based detection algorithms due to small sample sizes and class imbalance. Objective Given the limited number of actual CT reconstruction images of components and the sparse distribution of defects, this study aims to address the challenges of small sample sizes and class imbalance in defect detection model training by generating approximate CT reconstruction images to augment the defect detection training dataset. Methods We propose a novel CT detection image generation algorithm called the Decompound Synthesize Method (DSM), which decomposes the image generation process into three steps: model conversion, background generation, and defect synthesis. First, STL files of various industrial components are converted into voxel data, which undergo forward projection and image reconstruction to obtain corresponding CT images. Next, the Contour-CycleGAN model is employed to generate synthetic images that closely resemble actual CT images. Finally, defects are randomly sampled from an existing defect library and added to the images using the Copy-Adjust-Paste (CAP) method. These steps significantly expand the training dataset with images that closely mimic actual CT reconstructions. Results Experimental results validate the effectiveness of the proposed image generation method in defect detection tasks. Datasets generated using DSM exhibit greater similarity to actual CT images, and when combined with original data for training, these datasets enhance defect detection accuracy compared to using only the original images. Conclusion The DSM shows promise in addressing the challenges of small sample sizes and class imbalance. Future research can focus on further optimizing the generation algorithm and refining the model structure to enhance the performance and accuracy of defect detection models.

show abstract

A study on CT detection image generation based on decompound synthesize method

Fu,

Liu,

Zeng

et al. 2024

Journal of X-Ray Science and Technology: Clinical Applications

View full text Add to dashboard Cite

show abstract

Determination of Crack Depth in Brickworks by Ultrasonic Methods: Numerical Simulation and Regression Analysis

Beskopylny,

Stel’makh,

Shcherban’

et al. 2024

J. Compos. Sci.

View full text Add to dashboard Cite

Ultrasonic crack detection is one of the effective non-destructive methods of structural health monitoring (SHM) of buildings and structures. Despite its widespread use, crack detection in porous and heterogeneous composite building materials is an insufficiently studied issue and in practice leads to significant errors of more than 40%. The purpose of this article is to study the processes occurring in ceramic bricks weakened by cracks under ultrasonic exposure and to develop a method for determining the crack depth based on the characteristics of the obtained ultrasonic response. At the first stage, the interaction of the ultrasonic signal with the crack and the features of the pulse propagation process in ceramic bricks were considered using numerical modeling with the ANSYS environment. The FEM model allowed us to identify the characteristic aspects of wave propagation in bricks and compare the solution with the experimental one for the reference sample. Further experimental studies were carried out on ceramic bricks, as the most common elements of buildings and structures. A total of 110 bricks with different properties were selected. The cracks were natural or artificially created and were of varying depth and width. The experimental data showed that the greatest influence on the formation of the signal was exerted by the time parameters of the response: the time when the signal reaches a value of 12 units, the time of reaching the first maximum, the time of reaching the first minimum, and the properties of the material. Based on the regression analysis, a model was obtained that relates the crack depth to the signal parameters and the properties of the material. The error in the predicted values according to this model was approximately 8%, which was significantly more accurate than the existing approach.

show abstract

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

Li,

Gao,

Luo

et al. 2024

Sensors

View full text Add to dashboard Cite

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.

show abstract

Deep convolutional neural network for weld defect classification in radiographic images

Cited by 4 publications

References 75 publications

A study on CT detection image generation based on decompound synthesize method

A study on CT detection image generation based on decompound synthesize method

Determination of Crack Depth in Brickworks by Ultrasonic Methods: Numerical Simulation and Regression Analysis

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

Contact Info

Product

Resources

About