Ano-SuPs: Multisize Anomaly Detection for Manufactured Products by Identifying Suspected Patches via Vision Transformer

Image-based systems have gained popularity owing to their capacity to provide rich manufacturing status information, low implementation costs, and high acquisition rates. However, the complexity of the image background and various anomaly patterns pose new challenges to existing matrix-decomposition methods, which are inadequate for modeling requirements. Moreover, the uncertainty of the anomaly can cause anomaly-contamination problems, making the designed model and method highly susceptible to external disturbances. To address these challenges, we propose a two-step strategy anomaly-detection method that detects anomalies by identifying suspected patches. Specifically, we propose to detect the patches with anomalies by reconstructing the input image twice: the first step is to obtain a set of normal patches by removing those suspected patches, and the second step is to use those normal patches to refine the identification of the patches with anomalies. To demonstrate its effectiveness, we evaluate the proposed method systematically through simulation experiments and case studies. We further identify the key parameters and design steps that impact the model’s performance and efficiency.

History: Bianca Maria Colosimo served as the senior editor for this article.

Funding: This work was supported by the National Natural Science Foundation of China [Grant 72371219] and Guangdong Project [Grant 2024TQ08A432].