A Novel Pixel-Wise Defect Inspection Method Based on Stable Background Reconstruction
A Novel Pixel-Wise Defect Inspection Method Based on Stable Background Reconstruction
复制标题
DOI：
10.1109/tim.2020.3038413
复制
发表时间：
2020-11
影响因子：
5.6
通讯作者：
Chengkan Lv;Fei Shen;Zhengtao Zhang;De Xu;Yonghao He
中科院分区：
文献类型：
--
作者：
Chengkan Lv;Fei Shen;Zhengtao Zhang;De Xu;Yonghao He
本文提出一种基于背景重构的异常检测方法，对工业产品的纹理表面进行缺陷检测。该方法由两个模块组成：1）利用与生成对抗网络集成的自动编码器来重建原始图像的纹理背景作为无缺陷参考。具体来说，引入了额外的异常图像，并给出了异常的映射方法，以提高重建的稳定性。 2) 基于 U-net 的检查网络经过训练，可以对原始图像和重建的无缺陷图像之间的差异进行逐像素分析。在这些过程中，仅利用人工合成的缺陷图像来训练模型，而没有任何真实的缺陷样本。在多个纹理图像数据集和工业生产线上进行了一系列实验。实验结果表明了该方法的有效性和通用性。
In this article, an anomaly detection method based on background reconstruction is proposed to perform defect inspection on the texture surface of the industrial products. This method consists of two modules: 1) an autoencoder integrated with a generative adversarial network is utilized to reconstruct the textured background of the original image as a defect-free reference. Specifically, extra anomalous images are introduced and a mapping method of anomaly is given to improve the stability of reconstruction. 2) A U-net based inspection network is trained to perform pixel-wise analysis of the differences between the original and the reconstructed defect-free image. During these processes, only artificial synthesized defective images are utilized to train the model without any real defective samples. A series of experiments are conducted on several texture image data sets and the industrial production line. The experimental results reveal the effectiveness and versatility of the proposed method.