SPACE: SPAtial-aware Consistency rEgularization for anomaly detection in Industrial applications

TL;DR

SPACE introduces a spatial-aware regularization approach for anomaly detection in industrial settings, combining feature encoding, data augmentation, and regularization to improve detection accuracy and robustness.

Contribution

The paper presents a novel anomaly detection method, SPACE, integrating spatial consistency regularization and feature conversion to enhance detection in industrial applications.

Findings

Outperforms state-of-the-art methods on MVTec LOCO, MVTec AD, and VisA datasets.

Demonstrates improved detection accuracy and efficiency.

Effectively prevents overfitting and ambiguous feature learning.

Abstract

In this paper, we propose SPACE, a novel anomaly detection methodology that integrates a Feature Encoder (FE) into the structure of the Student-Teacher method. The proposed method has two key elements: Spatial Consistency regularization Loss (SCL) and Feature converter Module (FM). SCL prevents overfitting in student models by avoiding excessive imitation of the teacher model. Simultaneously, it facilitates the expansion of normal data features by steering clear of abnormal areas generated through data augmentation. This dual functionality ensures a robust boundary between normal and abnormal data. The FM prevents the learning of ambiguous information from the FE. This protects the learned features and enables more effective detection of structural and logical anomalies. Through these elements, SPACE is available to minimize the influence of the FE while integrating various data augmentations.In this study, we evaluated the proposed method on the MVTec LOCO, MVTec AD, and VisA datasets. Experimental results, through qualitative evaluation, demonstrate the superiority of detection and efficiency of each module compared to state-of-the-art methods.