Abstract:In the industrial scenario, anomaly detection could locate but cannot classify anomalies. To complete their capability, we study to automatically discover and recognize visual classes of industrial anomalies. In terms of multi-class anomaly classification, previous methods cluster anomalies represented by frozen pre-trained models but often fail due to poor discrimination. Novel class discovery (NCD) has the potential to tackle this. However, it struggles with non-prominent and semantically weak anomalies that challenge network learning focus. To address these, we introduce AnomalyNCD, a multi-class anomaly classification framework compatible with existing anomaly detection methods. This framework learns anomaly-specific features and classifies anomalies in a self-supervised manner. Initially, a technique called Main Element Binarization (MEBin) is first designed, which segments primary anomaly regions into masks to alleviate the impact of incorrect detections on learning. Subsequently, we employ mask-guided contrastive representation learning to improve feature discrimination, which focuses network attention on isolated anomalous regions and reduces the confusion of erroneous inputs through re-corrected pseudo labels. Finally, to enable flexible classification at both region and image levels during inference, we develop a region merging strategy that determines the overall image category based on the classified anomaly regions. Our method outperforms the state-of-the-art works on the MVTec AD and MTD datasets. Compared with the current methods, AnomalyNCD combined with zero-shot anomaly detection method achieves a 10.8% $F_1$ gain, 8.8% NMI gain, and 9.5% ARI gain on MVTec AD, 12.8% $F_1$ gain, 5.7% NMI gain, and 10.8% ARI gain on MTD. The source code is available at <a class="link-external link-https" href="https://github.com/HUST-SLOW/AnomalyNCD" rel="external noopener nofollow">this https URL</a>.

A Classifier-Based Approach to Multi-Class Anomaly Detection Applied to Astronomical Time-Series

A Classifier-Based Approach to Multi-Class Anomaly Detection for Astronomical Transients

Multi-Class Deep SVDD: Anomaly Detection Approach in Astronomy with Distinct Inlier Categories

Real-time Detection of Anomalies in Multivariate Time Series of Astronomical Data

Real-Time Detection of Anomalies in Large-Scale Transient Surveys

From Chaos to Clarity: Time Series Anomaly Detection in Astronomical Observations

AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios

Attention Based CNN-LSTM Network for Anomaly Pattern Classification of Multivariate Time Series

Learning to Detect Interesting Anomalies

Transformation Based Deep Anomaly Detection in Astronomical Images

Finding radio transients with anomaly detection and active learning based on volunteer classifications

Astronomaly at scale: searching for anomalies amongst 4 million galaxies

Intrinsic Anomaly Detection for Multi-Variate Time Series

A Multi-scale Parallel Unsupervised Model for Multivariate Time Series Anomaly Detection

Detecting Anomalous Images in Astronomical Datasets

Real-bogus scores for active anomaly detection

A Unified Model for Multi-class Anomaly Detection

MAG: A Novel Approach for Effective Anomaly Detection in Spacecraft Telemetry Data

Exploring the Universe with SNAD: Anomaly Detection in Astronomy

NLP Based Anomaly Detection for Categorical Time Series