Industrial Image Anomaly Detection Method Based on Improved MAE

Hui He, Yuanping Xu, Chaolong Zhang, Benjun Guo, Zhijie Xu, Jin Jin, Chao Kong, Jian Huang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


This study explores an improved Masked Autoencoder (MAE) model for anomaly detection in common industrial scene images. Industrial image anomaly detection, as an important research topic in computer vision, can detect abnormal data deviations from normal expected behavior and ensure the normal operation of various systems. In actual industrial scenarios, the scarcity of abnormal samples, the cost of labeled data, and the lack of prior knowledge about anomalies make unsupervised learning methods widely used in the field of image anomaly detection. However, most unsupervised learning methods currently require the application of large-scale datasets. In order to make unsupervised learning applicable to small-scale industrial image datasets, we replaced the backbone network with Swin Transformer based on the Masked Autoencoder method, while improving its mask strategy and optimizer. The effectiveness and advantage of the proposed method are demonstrated through an experimental comparison with existing model, the results show that this method is better than MAE model pretrained on ImageNet, and outperforms other unsupervised learning models in multiple categories.

Original languageEnglish
Title of host publication2023 28th International Conference on Automation and Computing (ICAC)
Number of pages6
ISBN (Electronic)9798350335859
ISBN (Print)9798350335866
Publication statusPublished - 16 Oct 2023
Event28th International Conference on Automation and Computing: Digitalisation for Smart Manufacturing and Systems - Aston University, Birmingham, United Kingdom
Duration: 30 Aug 20231 Sep 2023
Conference number: 28


Conference28th International Conference on Automation and Computing
Abbreviated titleICAC 2023
Country/TerritoryUnited Kingdom
Internet address

Cite this