Qualification Testing of Rebar Electric Arc Pressure Welding Joints Based on CNN-SVM for Apparent Features

DOI

10.2991/978-94-6463-449-5_59How to use a DOI?

Keywords

CNN-SVM; Electric arc pressure welding; Tensile strength

Abstract

This paper presents a convolutional neural network (CNN) and support vector machine (SVM) based analysis model for apparent features of electric arc pressure welding joints, aiming to enhance detection accuracy and replace traditional manual visual inspection methods. The model utilizes a training set of images of rebar welding joints processed by CNN to extract apparent feature vectors through denoising, matrix transformation, convolutional operations, and pooling steps. Simultaneously, the rebar undergoes tensile strength experiments, categorized as qualified and unqualified products. The extracted feature vectors are inputted into the SVM model, establishing a binary classification function model with the tensile strength results, and training the parameters of the CNN-SVM model. Finally, the welding joint test set is inputted into this model for inspection to observe the detection performance. The study demonstrates that the model achieves an accuracy of over 0.95, significantly higher than manual inspection, showing notable advantages.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zhiguo Yin
PY  - 2024
DA  - 2024/06/30
TI  - Qualification Testing of Rebar Electric Arc Pressure Welding Joints Based on CNN-SVM for Apparent Features
BT  - Proceedings of the 2024 8th International Conference on Civil Architecture and Structural Engineering (ICCASE 2024)
PB  - Atlantis Press
SP  - 604
EP  - 612
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-449-5_59
DO  - 10.2991/978-94-6463-449-5_59
ID  - Yin2024
ER  -