Unsupervised Anomaly Detection and Localization Based on Two-Hierarchy Normalizing Flow

Unsupervised anomaly detection (UAD) methods are widely used in industrial anomaly detection, primarily since there is a lack of anomalous data available for training. However, these methods still struggle to effectively detect and localize anomalies due to the diverse types of anomalies and frequent variations in their sizes and shapes. To address this challenge, we propose a new flow-based framework called two-hierarchy normalizing flow (THF) for anomaly detection and localization. THF consists of two main components: a masked flow and a constant flow. In the masked flow, we improve the traditional convolutional kernel by masking partial regions of the kernel to prevent excessive minimization of negative log-likelihood. This enhancement greatly helps in effectively detecting and localizing anomalies. The constant flow incorporates dual units (DUs) and a volume-preserving flow (VPF) module. The DUs consist of a characterization unit and a mixture unit. The characterization unit accurately captures both local and global feature information, while the mixture unit learns neighboring feature representations. Unlike traditional flow-based methods, the VPF module maintains the volume of the probability distribution invariable, facilitating precise detection and localization of high-scoring anomalies. Extensive experiments conducted on multiple widely used anomaly detection datasets demonstrate that THF outperforms state-of-the-art (SOTA) methods in both anomaly detection and localization.