MIAD: A Maintenance Inspection Dataset for Unsupervised Anomaly Detection

TL;DR

This paper introduces MIAD, a large-scale dataset for outdoor maintenance inspection anomaly detection, addressing the lack of suitable datasets for real-world outdoor scenarios with diverse anomalies and challenging conditions.

Contribution

The paper presents MIAD, a comprehensive, high-resolution dataset generated via 3D graphics, specifically designed for unsupervised anomaly detection in outdoor maintenance inspections, filling a critical gap in existing datasets.

Findings

Extensive evaluation of anomaly detection algorithms on MIAD.

MIAD covers diverse outdoor scenarios with surface and logical anomalies.

The dataset facilitates future research in outdoor unsupervised anomaly detection.

Abstract

Visual anomaly detection plays a crucial role in not only manufacturing inspection to find defects of products during manufacturing processes, but also maintenance inspection to keep equipment in optimum working condition particularly outdoors. Due to the scarcity of the defective samples, unsupervised anomaly detection has attracted great attention in recent years. However, existing datasets for unsupervised anomaly detection are biased towards manufacturing inspection, not considering maintenance inspection which is usually conducted under outdoor uncontrolled environment such as varying camera viewpoints, messy background and degradation of object surface after long-term working. We focus on outdoor maintenance inspection and contribute a comprehensive Maintenance Inspection Anomaly Detection (MIAD) dataset which contains more than 100K high-resolution color images in various outdoor industrial scenarios. This dataset is generated by a 3D graphics software and covers both surface and logical anomalies with pixel-precise ground truth. Extensive evaluations of representative algorithms for unsupervised anomaly detection are conducted, and we expect MIAD and corresponding experimental results can inspire research community in outdoor unsupervised anomaly detection tasks. Worthwhile and related future work can be spawned from our new dataset.