Self-Supervised Texture Image Anomaly Detection By Fusing Normalizing Flow and Dictionary Learning

TL;DR

This paper introduces a novel two-stage anomaly detection method that combines normalizing flow and dictionary learning, significantly improving detection accuracy for industrial texture images, especially in challenging cases.

Contribution

It presents a new approach integrating normalizing flow with dictionary learning to enhance texture anomaly detection beyond existing methods.

Findings

Achieved over 95% detection accuracy on MVTec AD texture data

Improved baseline accuracy from 67.9% to 99.7% on Carpet data

Demonstrated strong robustness across various industrial textures

Abstract

A common study area in anomaly identification is industrial images anomaly detection based on texture background. The interference of texture images and the minuteness of texture anomalies are the main reasons why many existing models fail to detect anomalies. We propose a strategy for anomaly detection that combines dictionary learning and normalizing flow based on the aforementioned questions. The two-stage anomaly detection approach already in use is enhanced by our method. In order to improve baseline method, this research add normalizing flow in representation learning and combines deep learning and dictionary learning. Improved algorithms have exceeded 95$\%$ detection accuracy on all MVTec AD texture type data after experimental validation. It shows strong robustness. The baseline method's detection accuracy for the Carpet data was 67.9%. The article was upgraded, raising the detection accuracy to 99.7%.