URA-Net: Uncertainty-Integrated Anomaly Perception and Restoration Attention Network for Unsupervised Anomaly Detection

2026-03-24

AI Summary

This work proposes URA-Net to address the challenge in unsupervised anomaly detection where conventional reconstruction-based methods often fail to effectively identify defects due to excessive generalization. URA-Net explicitly restores anomalous features to normal patterns through feature-level synthetic anomalies, a Bayesian neural network-driven uncertainty-aware module, and a global semantic-guided restoration attention mechanism, enabling precise localization via residual maps. By innovatively integrating uncertainty modeling with an anomaly restoration framework, the method overcomes the limitation of solely reconstructing normal samples. Extensive experiments on the MVTec AD, BTAD, and OCT-2017 datasets demonstrate that URA-Net significantly outperforms state-of-the-art approaches, validating its effectiveness and superiority in both industrial and medical image anomaly detection.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Adversarial Attacks & Robustness
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Unsupervised anomaly detection plays a pivotal role in industrial defect inspection and medical image analysis, with most methods relying on the reconstruction framework. However, these methods may suffer from over-generalization, enabling them to reconstruct anomalies well, which leads to poor detection performance. To address this issue, instead of focusing solely on normality reconstruction, we propose an innovative Uncertainty-Integrated Anomaly Perception and Restoration Attention Network (URA-Net), which explicitly restores abnormal patterns to their corresponding normality. First, unlike traditional image reconstruction methods, we utilize a pre-trained convolutional neural network to extract multi-level semantic features as the reconstruction target. To assist the URA-Net learning to restore anomalies, we introduce a novel feature-level artificial anomaly synthesis module to generate anomalous samples for training. Subsequently, a novel uncertainty-integrated anomaly perception module based on Bayesian neural networks is introduced to learn the distributions of anomalous and normal features. This facilitates the estimation of anomalous regions and ambiguous boundaries, laying the foundation for subsequent anomaly restoration. Then, we propose a novel restoration attention mechanism that leverages global normal semantic information to restore detected anomalous regions, thereby obtaining defect-free restored features. Finally, we employ residual maps between input features and restored features for anomaly detection and localization. The comprehensive experimental results on two industrial datasets, MVTec AD and BTAD, along with a medical image dataset, OCT-2017, unequivocally demonstrate the effectiveness and superiority of the proposed method.

Problem

• unsupervised anomaly detection
• over-generalization
• anomaly reconstruction
• industrial defect inspection
• medical image analysis

Innovation

• Uncertainty Integration
• Anomaly Restoration
• Bayesian Neural Networks
• Artificial Anomaly Synthesis
• Restoration Attention