2407.09578

Total: 1

#1 Unsupervised Anomaly Detection Using Diffusion Trend Analysis for Display Inspection [PDF5] [Copy] [Kimi2] [REL]

Authors: Eunwoo Kim, Un Yang, Cheol Lae Roh, Stefano Ermon

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.

Subjects: Computer Vision and Pattern Recognition , Machine Learning

Publish: 2024-07-12 01:50:07 UTC