Dual Attention-Based Industrial Surface Defect Detection with Consistency Loss

Abstract

In industrial production, flaws and defects inevitably appear on surfaces, resulting in unqualified products. Therefore, surface defect detection plays a key role in ensuring industrial product quality and maintaining industrial production lines. However, surface defects on different products have different manifestations, so it is difficult to regard all defective products as being within one category that has common characteristics. Defective products are also often rare in industrial production, making it difficult to collect enough samples. Therefore, it is appropriate to view the surface defect detection problem as a semi-supervised anomaly detection problem. In this paper, we propose an anomaly detection method that is based on dual attention and consistency loss to accomplish the task of surface defect detection. At the reconstruction stage, we employed both channel attention and pixel attention so that the network could learn more robust normal image reconstruction, which could in turn help to separate images of defects from defect-free images. Moreover, we proposed a consistency loss function that could exploit the differences between the multiple modalities of the images to improve the performance of the anomaly detection. Our experimental results showed that the proposed method could achieve a superior performance compared to the existing anomaly detection-based methods using the Magnetic Tile and MVTec AD datasets.

FEGAN: A Feature Extraction Based Approach for GAN Anomaly Detection and Localization

“…To quantitatively analyze the detection and segmentation performance of the FEGAN on the MVTec and Bottle-Cap, we compare the proposed method with seven alternative methods, namely GeoTrans [40], GANomaly [41], ITAE [42], DAGAN [43], ST-m [34], DFR [44], and SCGAN [38]. The anomaly detection results are presented in Table 1.…”

SCGAN: Extract Features From Normal Semantics for Unsupervised Anomaly Detection

“…Table 1 presents the testing outcomes of AnoGAN [3], GANomaly [25], Skip-GANomaly [36], DAGAN [20], CBiGAN [15], Dual-AttentionGAN [37], and SCGAN on the MVTec dataset. AUC data for AnoGAN, GANomaly, Skip-GANomaly, and DAGAN are sourced from the literature [20].…”

Anomaly Detection of GAN Industrial Image Based on Attention Feature Fusion

“…The results of the anomaly detection based on AFF are shown in Table 2 . We compared these with AnoGAN [ 7 ], GANomaly [ 13 ], Skip-GANomaly [ 15 ], DAGAN [ 21 ], CBiGAN [ 25 ] and Dual-AttentionGAN [ 20 ]. Among those, the AUC data in AnoGAN, GANomaly, Skip-GANomaly and DAGAN are taken from the literature [ 21 ].…”

“…In the test phase, in theory, the attribute restoration value of anomaly images should be very big. Li et al [ 20 ] proposed an anomaly detection method based on dual attention and consistency loss; the multiple scale channel attention and pixel attention are jointly used in the generative network based on an auto-encoder. Furthermore, pixel consistency, construction consistency and gradient consistency are added to the object function in order to enhance detail information retention.…”