Issue |
E3S Web Conf.
Volume 430, 2023
15th International Conference on Materials Processing and Characterization (ICMPC 2023)
|
|
---|---|---|
Article Number | 01252 | |
Number of page(s) | 11 | |
DOI | https://doi.org/10.1051/e3sconf/202343001252 | |
Published online | 06 October 2023 |
Computer Vision Algorithm for Predicting the Welding Efficiency of Friction Stir Welded Copper Joints from its Microstructures
1 School of Industrial and Information Engineering, Politecnico Di Milano, Milan, Italy
2 Symbiosis Institute of Technology, Symbiosis International (Deemed) University, Pune 412115, Maharashtra, India
3 Department of Finance, IIM Amritsar, India
4 Department of Material Science, Christian-Albrechts-University Kiel, Germany
* Corresponding author: vijaykumar.jatti@sitpune.edu.in
This research paper presents a study of the prediction of Friction Stir Welded (FSW) joint effectiveness using microstructure images with the aid of Convolutional Neural Networks (CNNs). A total of 3000 microstructure pictures were used for training the CNN, and 300 new microstructure photographs were used to test the accuracy of the model. The results showed that the CNN was able to accurately predict the effectiveness of FSW joints with an accuracy of 81 percent. The current work highlights the potential of using microstructure images and CNNs for improving the quality control and assessment of FSW joints in the materials and manufacturing industries. The findings of this study have important implications for the development of new techniques for improving the performance of FSW joints and for the wider application of computer vision and machine learning in the materials and manufacturing industries.
Key words: Computer Vision / Artificial Intelligence / Friction Stir Welding / Convolutional Neural Network
© The Authors, published by EDP Sciences, 2023
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.