Assisting Weld Defect Detection Using U2Net: A Machine Learning Approach.
Saved in:
| Title: | Assisting Weld Defect Detection Using U2Net: A Machine Learning Approach. |
|---|---|
| Authors: | Thompson, Cole J.1 (AUTHOR) cjthompson@lanl.gov, Porter, Reid B.1 (AUTHOR), O'Neil, Brian E.1 (AUTHOR), Charlton, William S.2 (AUTHOR) |
| Source: | Research in Nondestructive Evaluation. Sep-Dec2025, Vol. 36 Issue 5/6, p243-260. 18p. |
| Subjects: | Deep learning, Welding inspection, Transformer models, Machine learning, Convolutional neural networks, Radiography, Image segmentation |
| Abstract: | Weld defect detection is usually performed by expert human interpretation of radiography images. This process is time consuming and difficult. Deep learning solutions have been developed to assist interpretation. These solutions often use a U-Net convolutional neural network to distinguish between defect and non-defect pixels in the image. Transformers, like those used for ChatGPT, are widely adapted for computer vision tasks. This work describes a UNet, called U2Net, using transformer and convolutional U-Nets to encode global and local information simultaneously for weld defect segmentation in radiographs. Results show that U2Net provides beneficial defect detection capabilities as compared to four evaluated U-Nets, underscoring the benefit of coupling global and local information in weld defect detection. [ABSTRACT FROM AUTHOR] |
| Copyright of Research in Nondestructive Evaluation is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
Be the first to leave a comment!
