DRL-Guided Neural Batch Sampling for Semi-Supervised Pixel-Level Anomaly Detection

TL;DR

This paper introduces a semi-supervised reinforcement learning approach for pixel-level anomaly detection in industrial images, effectively utilizing limited labeled data to improve defect localization and detection accuracy.

Contribution

It presents a novel RL-guided neural batch sampling framework that enhances semi-supervised anomaly detection by adaptively selecting informative patches for training.

Findings

Achieves higher detection accuracy than recent state-of-the-art methods.

Improves localization of subtle anomalies in industrial images.

Maintains low complexity with significant performance gains.

Abstract

Anomaly detection in industrial visual inspection is challenging due to the scarcity of defective samples. Most existing methods rely on unsupervised reconstruction using only normal data, often resulting in overfitting and poor detection of subtle defects. We propose a semi-supervised deep reinforcement learning framework that integrates a neural batch sampler, an autoencoder, and a predictor. The RL-based sampler adaptively selects informative patches by balancing exploration and exploitation through a composite reward. The autoencoder generates loss profiles highlighting abnormal regions, while the predictor performs segmentation in the loss-profile space. This interaction enables the system to effectively learn both normal and defective patterns with limited labeled data. Experiments on the MVTec AD dataset demonstrate that our method achieves higher accuracy and better localization of subtle anomalies than recent state-of-the-art approaches while maintaining low complexity, yielding an average improvement of 0.15 in F1_max and 0.06 in AUC, with a maximum gain of 0.37 in F1_max in the best case.