Unsupervised Anomaly Detection Using Diffusion Trend Analysis for Display Inspection
This work addresses practical challenges in display inspection for manufacturing or quality control, but it appears incremental as it builds on existing diffusion models.
The paper tackled the problem of reconstruction-based anomaly detection in display inspection, where existing denoising diffusion models suffer from issues with noise parameter selection and false detection due to normal region fluctuations; the proposed method analyzes reconstruction trends to effectively solve these problems, though no concrete numbers are provided in the abstract.
Reconstruction-based anomaly detection via denoising diffusion model has limitations in determining appropriate noise parameters that can degrade anomalies while preserving normal characteristics. Also, normal regions can fluctuate considerably during reconstruction, resulting in false detection. In this paper, we propose a method to detect anomalies by analysis of reconstruction trend depending on the degree of degradation, effectively solving the both problems that impede practical application in display inspection.