Positive-Sample-Based Surface Defect Detection Using Memory-Augmented Adversarial Autoencoders

In industrial quality inspection, large amounts of data on the desired product appearance are available at the time of training, while significantly few defective samples are available. In this study, we proposed new memory-augmented adversarial autoencoders to detect and localize defects in real-time using defect-free samples alone for model training. This research was conducted by reconstructing images using an adversarial autoencoder and detection results from the Frechet Markov distance (FMD). A threshold was determined based on the statistical characteristics of defect-free samples in the training set. Innovatively, we introduced a memory module and redesigned the reconstruction loss function to avoid the situation where the reconstruction ability is too strong or too poor, which lead to missing detection of defects. Then we proposed FMD, which can accurately measure the distance between the distribution of test samples and positive samples. Moreover, the statistics-based threshold determination method is used to meet different industrial needs. The accuracy, robustness, and computational overhead of the proposed MAA were evaluated using three datasets obtained from the production line and two benchmark datasets. The results indicated the effectiveness and ability of the proposed method to adapt to the real-time nature of industrial production.

Automated summary

In this study, a new memory-augmented adversarial autoencoder (MAA) is proposed for real-time defect detection and localization using defect-free samples only. By introducing a memory module and redesigning the reconstruction loss function, the proposed MAA avoids missing defect detection. The results show that the proposed method is effective, accurate, and robust, suitable for real-time industrial production.