HSS-IAD: A Heterogeneous Same-Sort Industrial Anomaly Detection Dataset

2025-04-17

AI Summary

Existing industrial anomaly detection (IAD) datasets suffer from category mixing, insufficient coverage of structural/appearance variations, and unrealistic defect distortions, undermining the generalizability of multi-class unsupervised anomaly detection (MUAD) algorithms in real production lines. Method: We introduce MIDB—the first IAD benchmark dedicated to same-category metal components—featuring a novel “same-category heterogeneous” data paradigm that captures structural and appearance diversity, material-level subtle defects, and provides high-precision pixel-level annotations alongside a foreground-mask-supported synthetic anomaly generation toolkit. Contribution/Results: Based on MIDB, we establish a comprehensive MUAD evaluation framework incorporating both multi-class joint and single-class isolated settings. Evaluating 12 state-of-the-art methods on 8,580 images, we rigorously expose their performance bottlenecks under realistic industrial conditions, thereby filling a critical gap in production-line adaptability assessment and advancing algorithmic alignment with industrial deployment requirements.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Multi-class Unsupervised Anomaly Detection algorithms (MUAD) are receiving increasing attention due to their relatively low deployment costs and improved training efficiency. However, the real-world effectiveness of MUAD methods is questioned due to limitations in current Industrial Anomaly Detection (IAD) datasets. These datasets contain numerous classes that are unlikely to be produced by the same factory and fail to cover multiple structures or appearances. Additionally, the defects do not reflect real-world characteristics. Therefore, we introduce the Heterogeneous Same-Sort Industrial Anomaly Detection (HSS-IAD) dataset, which contains 8,580 images of metallic-like industrial parts and precise anomaly annotations. These parts exhibit variations in structure and appearance, with subtle defects that closely resemble the base materials. We also provide foreground images for synthetic anomaly generation. Finally, we evaluate popular IAD methods on this dataset under multi-class and class-separated settings, demonstrating its potential to bridge the gap between existing datasets and real factory conditions. The dataset is available at https://github.com/Qiqigeww/HSS-IAD-Dataset.

Problem

• Limitations in current Industrial Anomaly Detection datasets
• Defects not reflecting real-world characteristics
• Need for dataset with varied structures and subtle defects

Innovation

• Introduces HSS-IAD dataset with metallic-like parts
• Provides precise anomaly annotations and variations
• Evaluates IAD methods under real-world conditions