Noise-to-Norm Reconstruction for Industrial Anomaly Detection and Localization

TL;DR

This paper introduces a noise-to-norm reconstruction method for industrial anomaly detection that effectively reconstructs anomalies into normal patterns, improving detection and localization accuracy especially on datasets with large positional variations.

Contribution

It proposes a novel noise-to-norm reconstruction approach using an M-net with multiscale fusion and residual attention, enhancing anomaly detection and localization performance.

Findings

Achieved state-of-the-art results on the MPDD dataset.

Demonstrated superior anomaly localization accuracy.

Effective in reconstructing anomalous regions into normal patterns.

Abstract

Anomaly detection has a wide range of applications and is especially important in industrial quality inspection. Currently, many top-performing anomaly-detection models rely on feature-embedding methods. However, these methods do not perform well on datasets with large variations in object locations. Reconstruction-based methods use reconstruction errors to detect anomalies without considering positional differences between samples. In this study, a reconstruction-based method using the noise-to-norm paradigm is proposed, which avoids the invariant reconstruction of anomalous regions. Our reconstruction network is based on M-net and incorporates multiscale fusion and residual attention modules to enable end-to-end anomaly detection and localization. Experiments demonstrate that the method is effective in reconstructing anomalous regions into normal patterns and achieving accurate anomaly detection and localization. On the MPDD and VisA datasets, our proposed method achieved more competitive results than the latest methods, and it set a new state-of-the-art standard on the MPDD dataset.