Content area

Abstract

Conference Title: 2025 IEEE 15th International Conference on CYBER Technology in Automation, Control, and Intelligent Systems (CYBER)

Conference Start Date: 2025 July 15

Conference End Date: 2025 July 18

Conference Location: Shanghai, China

In industrial Wire Arc Additive Manufacturing (WAAM)., the scarcity and imbalance of labeled defect images limit the effectiveness of deep learning-based quality inspection systems. This paper presents a hybrid CVAE-CGAN framework designed to generate high-resolution., class-conditional molten pool images for data augmentation in small-sample settings. By combining a conditional variational autoencoder with adversarial training and integrating VGG19-based perceptual loss and sub-pixel convolution, the proposed model produces visually realistic and diverse synthetic defect images. Extensive experiments on a nine-class WAAM defect dataset demonstrate the model's ability to enhance classification performance, especially for underrepresented categories, offering a scalable solution to mitigate data limitations in intelligent manufacturing.

Details

Title
Generation of WAAM Defect Images Using a Hybrid CVAE-CGAN: A Data Augmentation Strategy for Small and Imbalanced Datasets
Author
Yang, Junle 1 ; Yuan, Lei 1 ; Mu, Haochen 2 ; He, Fengyang 1 ; Ding, Donghong 2 ; Pan, Zengxi 1 ; Li, Huijun 1 

 School of Mechanical, Materials, Mechatronic and Biomedical Engineering, University of Wollongong,Wollongong,NSW,Australia,2522 
 School of Mechanical and Power Engineering, Nanjing Tech University,Nanjing,China,211816 
Pages
1-6
Number of pages
6
Publication year
2025
Publication date
2025
Publisher
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Place of publication
Piscataway
Country of publication
United States
Source type
Conference Paper
Language of publication
English
Document type
Conference Proceedings
Publication history
 
 
Online publication date
2025-09-24
Publication history
 
 
   First posting date
24 Sep 2025
ProQuest document ID
3253885402
Document URL
https://www.proquest.com/conference-papers-proceedings/generation-waam-defect-images-using-hybrid-cvae/docview/3253885402/se-2?accountid=208611
Copyright
Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2025
Last updated
2025-09-25
Database
ProQuest One Academic