Component-aware anomaly detection framework for adjustable and logical industrial visual inspection

15 May 2023  ·  Tongkun Liu, Bing Li, Xiao Du, Bingke Jiang, Xiao Jin, Liuyi Jin, Zhuo Zhao ·

Industrial visual inspection aims at detecting surface defects in products during the manufacturing process. Although existing anomaly detection models have shown great performance on many public benchmarks, their limited adjustability and ability to detect logical anomalies hinder their broader use in real-world settings. To this end, in this paper, we propose a novel component-aware anomaly detection framework (ComAD) which can simultaneously achieve adjustable and logical anomaly detection for industrial scenarios. Specifically, we propose to segment images into multiple components based on a lightweight and nearly training-free unsupervised semantic segmentation model. Then, we design an interpretable logical anomaly detection model through modeling the metrological features of each component and their relationships. Despite its simplicity, our framework achieves state-of-the-art performance on image-level logical anomaly detection. Meanwhile, segmenting a product image into multiple components provides a novel perspective for industrial visual inspection, demonstrating great potential in model customization, noise resistance, and anomaly classification. The code will be available at https://github.com/liutongkun/ComAD.

PDF Abstract

Datasets


Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Anomaly Detection MVTec LOCO AD ComAD+DRAEM Avg. Detection AUROC 87.9 # 13
Detection AUROC (only logical) 85.9 # 15
Detection AUROC (only structural) 89.9 # 11
Anomaly Detection MVTec LOCO AD ComAD+RD4AD Avg. Detection AUROC 88.2 # 12
Detection AUROC (only logical) 87.5 # 12
Detection AUROC (only structural) 88.8 # 13
Anomaly Detection MVTec LOCO AD ComAD+AST Avg. Detection AUROC 89.8 # 10
Detection AUROC (only logical) 90.1 # 7
Detection AUROC (only structural) 89.4 # 12
Anomaly Detection MVTec LOCO AD ComAD+PatchCore Avg. Detection AUROC 90.1 # 8
Detection AUROC (only logical) 89.4 # 9
Detection AUROC (only structural) 90.9 # 9
Anomaly Detection MVTec LOCO AD ComAD Avg. Detection AUROC 81.2 # 25
Detection AUROC (only logical) 87.7 # 11
Detection AUROC (only structural) 74.6 # 31

Methods


No methods listed for this paper. Add relevant methods here