TY - JOUR

T1 - Masked Swin Transformer Unet for Industrial Anomaly Detection

AU - Jiang, Jielin

AU - Zhu, Jiale

AU - Bilal, Muhammad

AU - Cui, Yan

AU - Kumar, Neeraj

AU - Dou, Ruihan

AU - Su, Feng

AU - Xu, Xiaolong

PY - 2023/2/1

Y1 - 2023/2/1

N2 - The intelligent detection process for industrial anomalies employs artificial intelligence methods to classify images that deviate from a normal appearance. Traditional convolutional neural network (CNN)-based anomaly detection algorithms mainly use the network to restructure abnormal areas and detect anomalies by calculating the errors between the original image and reconstructed image. However, the traditional CNNs struggle to extract global context information, resulting in poor anomaly detection performance. Thus, a masked Swin Transformer Unet (MSTUnet) for anomaly detection is proposed. To solve the problem of insufficient abnormal samples in the training phase, an anomaly simulation and mask strategy is first applied on anomaly-free samples to generate a simulated anomaly and, then, the Swin Transformer's powerful global learning ability is used to inpaint the masked area. Finally, a convolution-based Unet network is used for end-to-end anomaly detection. Experimental results on industrial dataset MVTec AD show that MSTUnet achieves superior anomaly detection and localization performance.

AB - The intelligent detection process for industrial anomalies employs artificial intelligence methods to classify images that deviate from a normal appearance. Traditional convolutional neural network (CNN)-based anomaly detection algorithms mainly use the network to restructure abnormal areas and detect anomalies by calculating the errors between the original image and reconstructed image. However, the traditional CNNs struggle to extract global context information, resulting in poor anomaly detection performance. Thus, a masked Swin Transformer Unet (MSTUnet) for anomaly detection is proposed. To solve the problem of insufficient abnormal samples in the training phase, an anomaly simulation and mask strategy is first applied on anomaly-free samples to generate a simulated anomaly and, then, the Swin Transformer's powerful global learning ability is used to inpaint the masked area. Finally, a convolution-based Unet network is used for end-to-end anomaly detection. Experimental results on industrial dataset MVTec AD show that MSTUnet achieves superior anomaly detection and localization performance.

KW - Anomaly detection

KW - inpainting

KW - Swin Transformer

KW - Unet

U2 - 10.1109/TII.2022.3199228

DO - 10.1109/TII.2022.3199228

M3 - Journal article

AN - SCOPUS:85136891169

VL - 19

SP - 2200

EP - 2209

JO - IEEE Transactions on Industrial Informatics

JF - IEEE Transactions on Industrial Informatics

SN - 1551-3203

IS - 2

ER -