Estimating the Probability Density Function of New Fabrics for Fabric Anomaly Detection

Oliver Rippel, Maximilian Müller, Andreas Münkel, Thomas Gries, Dorit Merhof

Abstract

Image-based quality control aims at detecting anomalies (i.e. defects) in products. Supervised, data driven approaches have greatly improved Anomaly Detection (AD) performance, but suffer from a major drawback: they require large amounts of annotated training data, limiting their economic viability. In this work, we challenge and overcome this limitation for complex patterned fabrics. Investigating the structure of deep feature representations learned on a large-scale fabric dataset, we find that fabrics form clusters according to their fabric type, whereas anomalies form a cluster on their own. We leverage this clustering behavior to estimate the Probability Density Function (PDF) of new, previously unseen fabrics, in the deep feature representations directly. Using this approach, we outperform supervised and semi-supervised AD approaches trained on new fabrics, requiring only defect-free data for PDF-estimation.

Download


Paper Citation


in Harvard Style

Rippel O., Müller M., Münkel A., Gries T. and Merhof D. (2021). Estimating the Probability Density Function of New Fabrics for Fabric Anomaly Detection.In Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-486-2, pages 463-470. DOI: 10.5220/0010163604630470


in Bibtex Style

@conference{icpram21,
author={Oliver Rippel and Maximilian Müller and Andreas Münkel and Thomas Gries and Dorit Merhof},
title={Estimating the Probability Density Function of New Fabrics for Fabric Anomaly Detection},
booktitle={Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2021},
pages={463-470},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010163604630470},
isbn={978-989-758-486-2},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 10th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Estimating the Probability Density Function of New Fabrics for Fabric Anomaly Detection
SN - 978-989-758-486-2
AU - Rippel O.
AU - Müller M.
AU - Münkel A.
AU - Gries T.
AU - Merhof D.
PY - 2021
SP - 463
EP - 470
DO - 10.5220/0010163604630470