Progress in Welding Defect Detection Based on Visual Technology

DOI

10.2991/978-94-6239-648-7_41How to use a DOI?

Keywords

Weld defect detection; Machine vision; Deep learning; Feature extraction; Smart manufacturing

Abstract

Welding defect detection is a crucial step in ensuring the safety of industrial manufacturing. The traditional manual visual detection and radiographic detection methods have low efficiency, high costs, and poor adaptability, which make it challenging to meet the needs of large-scale manufacturing. With the development of computer vision and machine learning methods, welding defect detection methods based on vision have become a research hotspot because of their non-contact, high efficiency, and intelligence. This paper reviews the research progress in this field. To begin with, the application scenarios and limitations of several traditional detection methods are introduced, and their shortcomings in dealing with a strong interference environment and identifying minor defects are emphasized. Then, the basic principle of machine learning and its application in weld defect detection are analyzed, and the innovative application of machine learning and deep learning technology in weld defect detection is discussed, including the excellent performance of support vector machine, random forest, convolutional neural network, transformer, and other models in feature extraction and classification. In addition, this paper also summarizes the key challenges currently facing us, including suppressing interference under extreme conditions, overcoming the bottleneck of high-precision detection, and addressing the obstacles to industrial implementation. Future research directions are identified, including multimodal data fusion, real-time adaptive control, lightweight models, and improving standardization and interpretability. Through technology integration and scene refinement, vision-based welding defect detection may transition from “passive recognition” to “active prediction”, providing more efficient and accurate quality control solutions for intelligent manufacturing.

Copyright

© 2026 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  - Xucheng Feng
PY  - 2026
DA  - 2026/04/24
TI  - Progress in Welding Defect Detection Based on Visual Technology
BT  - Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
PB  - Atlantis Press
SP  - 370
EP  - 376
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-648-7_41
DO  - 10.2991/978-94-6239-648-7_41
ID  - Feng2026
ER  -