科学技术与应用

基于双判别器生成对抗网络的工业表面缺陷图像异常检测方法研究

王 红晨

陕西国际商贸学院

王 玉芳

陕西国际商贸学院

DOI: https://doi.org/10.59429/kxjsyy.v3i4.15189

Keywords: 异常检测；生成对抗网络；工业图像；表面缺陷；无监督学习

---

Abstract

针对工业质检场景下缺陷样本获取困难、类别分布严重失衡致使传统监督式深度模型泛化能力受限的问题， 本文提出一种基于双判别器生成对抗网络（Dual-Discriminator Generative Adversarial Network，DD-GAN）的图像异 常检测方法。模型训练阶段仅使用正常样本，通过引入面向整图语义的全局判别器与面向局部纹理的局部判别器协 同参与对抗训练，并将像素级重构损失、特征匹配损失与对抗损失联合优化，从而增强模型对微小缺陷区域的敏感 度。在推理阶段，本文构造融合像素差异与深层特征差异的复合异常评分函数，实现对未知缺陷类型的稳健检测。 基于 MVTec AD 公开数据集的多组对比实验与消融实验表明，所提方法在九类工业产品上的平均 AUC 达到 0.953， 较基线 GANomaly 方法提升 4.2 个百分点，验证了双判别器结构与复合评分策略在提升检测性能方面的有效性。

---

References

[1] Bergmann P, Fauser M, Sattlegger D, et al. MVTec AD—A comprehensive real-world dataset for unsupervised anomaly detection [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2019: 9592-9600.

[2] Goodfellow I, Pouget-Abadie J, Mirza M, et al.Generative adversarial networks [J]. Communications of the ACM, 2020, 63(11): 139-144.

[3] Schlegl T, Seeböck P, Waldstein S M, et al.Unsupervised anomaly detection with generative adversarial networks to guide marker discovery [C]// International

Conference on Information Processing in Medical Imaging.Springer, 2017: 146-157.

[4] Akcay S, Atapour-Abarghouei A, Breckon T P.GANomaly: Semi-supervised anomaly detection via adversarial training [C]// Asian Conference on Computer Vision. Springer,2018: 622-637.

[5] Akcay S, Atapour-Abarghouei A, Breckon T P. Skip-GANomaly: Skip connected and adversarially trained encoder-decoder anomaly detection [C]// International Joint Conference on Neural Networks. IEEE, 2019: 1-8.

[6] Schlegl T, Seeböck P, Waldstein S M, et al.f-AnoGAN: Fast unsupervised anomaly detection with generative adversarial networks [J]. Medical Image Analysis,2019, 54: 30-44.

[7] Defard T, Setkov A, Loesch A, et al. PaDiM: A patch distribution modeling framework for anomaly detection and localization [C]// International Conference on Pattern Recognition. Springer, 2021: 475-489.

[8] Roth K, Pemula L, Zepeda J, et al. Towards total recall in industrial anomaly detection [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022: 14318-14328.

[9] 陶显，侯伟，徐德. 基于深度学习的表面缺陷检测方法综述[J]. 自动化学报，2021, 47(5): 1017-1034.

[10] 陈健，张开生，谷全全. 工业产品表面缺陷视觉检测研究综述[J]. 计算机工程与应用，2022, 58(15): 1-17.

[11] Isola P, Zhu J Y, Zhou T, et al. Image-to-image translation with conditional adversarial networks [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017: 1125-1134.

[12] Mao X, Li Q, Xie H, et al. Least squares generative adversarial networks [C]// Proceedings of the IEEE International Conference on Computer Vision. 2017: 2794-2802.