A Novel Representation of Periodic Pattern and Its Application to Untrained Anomaly Detection

TL;DR

This paper introduces a new method for representing and detecting anomalies in periodic industrial textures, effectively handling noise and unknown anomalies through a joint optimization framework.

Contribution

It proposes a novel self-representation model for periodic patterns that integrates anomaly detection into a unified optimization process.

Findings

Effective periodic pattern learning demonstrated on simulated data

Enhanced anomaly detection performance on real-world industrial images

Robustness to noise and unknown anomalies in pattern extraction

Abstract

There are a variety of industrial products that possess periodic textures or surfaces, such as carbon fiber textiles and display panels. Traditional image-based quality inspection methods for these products require identifying the periodic patterns from normal images (without anomaly and noise) and subsequently detecting anomaly pixels with inconsistent appearances. However, it remains challenging to accurately extract the periodic pattern from a single image in the presence of unknown anomalies and measurement noise. To deal with this challenge, this paper proposes a novel self-representation of the periodic image defined on a set of continuous parameters. In this way, periodic pattern learning can be embedded into a joint optimization framework, which is named periodic-sparse decomposition, with simultaneously modeling the sparse anomalies and Gaussian noise. Finally, for the real-world industrial images that may not strictly satisfy the periodic assumption, we propose a novel pixel-level anomaly scoring strategy to enhance the performance of anomaly detection. Both simulated and real-world case studies demonstrate the effectiveness of the proposed methodology for periodic pattern learning and anomaly detection.