Scratch detection based on image reconstruction using ideal-image-pair dataset of just before/after defect occurrence

When inspecting defects such as scratches due to image processing, if we can obtain an image before defect occurrence, the defect can be detected by simply comparing the image pair before and after defect occurrence. However, this idea is generally unrealistic because it is impossible to obtain an image before defect occurrence from an image after the defect occurred unless we go back in time. Therefore, we propose a method of training a generation-based model to detect scratches based on subtraction between an input image and output image during testing. We obtained a large amount of ideal image pairs before and after defect occurrence (i.e., image pairs in which only defect regions are different and the others are almost completely the same) using an image-capturing device. An image with a background texture of high reconstruction performance was generated with our method by using this dataset. Although we used image pairs of only an aluminum plate for training, the area under the curve of the receiver operating characteristic (ROC-AUC) measure was 0.9973 for a copper plate and 0.9904 for a stainless steel plate. This shows that our method is also effective for background textures different from those used for training. We also compared our generation-based method with a method of training a classificationbased model (ResNet18). We confirmed that the ROC-AUC measure with the proposed method was 0.9824 for an acrylic plate and 0.6156 for a carbon fiber sheet, which is better than those with the classification-based method (0.9650 and 0.5000, respectively). The proposed method has much higher reconstruction performance than previous training of defect-free images only, and is effective for scratch detection of background texture similar to those used for training.

[1]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[2]  Carsten Steger,et al.  Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoders , 2018, VISIGRAPP.

[3]  Nassir Navab,et al.  Deep Autoencoding Models for Unsupervised Anomaly Segmentation in Brain MR Images , 2018, BrainLes@MICCAI.

[4]  Toby P. Breckon,et al.  Skip-GANomaly: Skip Connected and Adversarially Trained Encoder-Decoder Anomaly Detection , 2019, 2019 International Joint Conference on Neural Networks (IJCNN).

[5]  Georg Langs,et al.  Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery , 2017, IPMI.

[6]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[7]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[8]  Chuan Sheng Foo,et al.  Efficient GAN-Based Anomaly Detection , 2018, ArXiv.

[9]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  Chien-Fang Ding,et al.  Anomaly Detection Neural Network with Dual Auto-Encoders GAN and Its Industrial Inspection Applications , 2020, Sensors.

[11]  Diganta Misra.  Mish: A Self Regularized Non-Monotonic Activation Function , 2020, BMVC.

[12]  Alexei A. Efros,et al.  Image-to-Image Translation with Conditional Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[13]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.