To read this content please select one of the options below:

Defect detection of color-patterned fabric based on DenoisingGAN

Hongwei Zhang (School of Electronics and Information, Xi'an Polytechnic University, Xi'an, China)
Shihao Wang (School of Electronics and Information, Xi'an Polytechnic University, Xi'an, China)
Hongmin Mi (School of Electronics and Information, Xi'an Polytechnic University, Xi'an, China)
Shuai Lu (The Institute of Engineering Medicine, Beijing Institute of Technology, Beijing, China)
Le Yao (State Key Laboratory of Industrial Control Technology, Zhejiang University, Jinan, China)
Zhiqiang Ge (College of Control Science and Engineering, Zhejiang University, Jinan, China)

International Journal of Clothing Science and Technology

ISSN: 0955-6222

Article publication date: 31 August 2023

Issue publication date: 31 October 2023

175

Abstract

Purpose

The defect detection problem of color-patterned fabric is still a huge challenge due to the lack of manual defect labeling samples. Recently, many fabric defect detection algorithms based on feature engineering and deep learning have been proposed, but these methods have overdetection or miss-detection problems because they cannot adapt to the complex patterns of color-patterned fabrics. The purpose of this paper is to propose a defect detection framework based on unsupervised adversarial learning for image reconstruction to solve the above problems.

Design/methodology/approach

The proposed framework consists of three parts: a generator, a discriminator and an image postprocessing module. The generator is able to extract the features of the image and then reconstruct the image. The discriminator can supervise the generator to repair defects in the samples to improve the quality of image reconstruction. The multidifference image postprocessing module is used to obtain the final detection results of color-patterned fabric defects.

Findings

The proposed framework is compared with state-of-the-art methods on the public dataset YDFID-1(Yarn-Dyed Fabric Image Dataset-version1). The proposed framework is also validated on several classes in the MvTec AD dataset. The experimental results of various patterns/classes on YDFID-1 and MvTecAD demonstrate the effectiveness and superiority of this method in fabric defect detection.

Originality/value

It provides an automatic defect detection solution that is convenient for engineering applications for the inspection process of the color-patterned fabric manufacturing industry. A public dataset is provided for academia.

Keywords

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 61803292), the Key R&D plan of Shaanxi Province (No. 2019ZDLGY01-08), the Innovation Talent Promotion Foundation of Shaanxi Province (No. 2018KJXX-038), the Natural Science Foundation of Shaanxi Province (No. 2019JM-263), the Special scientific research plan project of Shaanxi Provincial Department of Education (No. 17JK0577), and the Innovation Capability Support Program of Shaanxi (No. 2021TD-29).

Since acceptance of this article, the following author(s) have updated their affiliation(s): Hongwei Zhang and Zhiqiang Ge are at the State Key Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou, China and Le Yao is at the School of Mathematics, Hangzhou Normal University, Hangzhou, China.

Erratum: It has come to the attention of the publisher that the article “Defect detection of color-patterned fabric based on DenoisingGAN”, International Journal of Clothing Science and Technology, Vol. 35 No. 6, pp. 865-888. https://doi.org/10.1108/IJCST-03-2022-0032, contained an error in the affiliation addresses for Le Yao and Zhiqiang Ge, which was introduced during the production process. ‘Le Yao (State Key Laboratory of Industrial Control Technology, Zhejiang University, Jinan, China)’ has been corrected to ‘Le Yao (State Key Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou, China)’ and ‘Zhiqiang Ge (College of Control Science and Engineering, Zhejiang University, Jinan, China)’ has been corrected to ‘Zhiqiang Ge (College of Control Science and Engineering, Zhejiang University, Hangzhou, China)’. The publisher sincerely apologises for this error and for any confusion caused.

Citation

Zhang, H., Wang, S., Mi, H., Lu, S., Yao, L. and Ge, Z. (2023), "Defect detection of color-patterned fabric based on DenoisingGAN", International Journal of Clothing Science and Technology, Vol. 35 No. 6, pp. 865-888. https://doi.org/10.1108/IJCST-03-2022-0032

Publisher

:

Emerald Publishing Limited

Copyright © 2023, Emerald Publishing Limited

Related articles