PatchProto Networks for Few-shot Visual Anomaly Classification

TL;DR

This paper introduces PatchProto networks, a novel few-shot learning approach that uses CNN features of defective regions for improved anomaly classification in industrial quality inspection.

Contribution

It proposes a new method focusing on defective region features, enhancing few-shot anomaly classification accuracy over traditional methods.

Findings

Significant improvement in classification accuracy on MVTec-AD dataset.

Effective use of defective region features for few-shot learning.

Outperforms basic few-shot classifiers in anomaly detection tasks.

Abstract

The visual anomaly diagnosis can automatically analyze the defective products, which has been widely applied in industrial quality inspection. The anomaly classification can classify the defective products into different categories. However, the anomaly samples are hard to access in practice, which impedes the training of canonical machine learning models. This paper studies a practical issue that anomaly samples for training are extremely scarce, i.e., few-shot learning (FSL). Utilizing the sufficient normal samples, we propose PatchProto networks for few-shot anomaly classification. Different from classical FSL methods, PatchProto networks only extract CNN features of defective regions of interest, which serves as the prototypes for few-shot learning. Compared with basic few-shot classifier, the experiment results on MVTec-AD dataset show PatchProto networks significantly improve the few-shot anomaly classification accuracy.