Surface defect detection method based on generative adversarial networks

Zifan Chai; Liqin Zhao; Haibo Jiang; Nengquan Duan; Qingyan Ma; Benbo Zhao

doi:10.1117/12.3034164

13 June 2024 Surface defect detection method based on generative adversarial networks

Zifan Chai, Liqin Zhao, Haibo Jiang, Nengquan Duan, Qingyan Ma, Benbo Zhao

Proceedings Volume 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024); 1318062 (2024) https://doi.org/10.1117/12.3034164
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 2024, Guangzhou, China

Abstract

Many existing defect detection algorithms are tailored to specific on-site conditions, utilizing single models. When confronted with defect detection scenarios across various settings, their generalization performance is often poor. Overly specific feature descriptions significantly limit the generalization capabilities of defect detection models. In this paper, We propose a surface defect detection method based on positive samples, which not only solves the problem of difficult model training due to the lack of negative samples in industrial sites, but also overcomes the challenges of poor model generalization and low detection accuracy in complex environments. Firstly, a bidirectional generative adversarial network with two discriminators is introduced to reconstruct images, thereby imposing constraints on the cycle consistency of the reconstructed images. Simultaneously, we propose a weighted anomaly score evaluation method using residual loss and discrimination loss to assess the differences between the reconstructed images and the original images, achieving defect detection in various scenarios.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Zifan Chai, Liqin Zhao, Haibo Jiang, Nengquan Duan, Qingyan Ma, and Benbo Zhao "Surface defect detection method based on generative adversarial networks", Proc. SPIE 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 1318062 (13 June 2024); https://doi.org/10.1117/12.3034164

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
9 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Image restoration

Defect detection

Education and training

Data modeling

Statistical modeling

Deep learning

Gallium nitride

Show All Keywords

Keywords/Phrases

Search In:

Publication Years