Increasing the Generalization of Supervised Fabric Anomaly Detection Methods to Unseen Fabrics
Open Access
Abstract
Fabric anomaly detection (AD) tries to detect anomalies (i.e., defects) in fabrics, and fabric AD approaches are continuously improved with respect to their AD performance. However, developed solutions are known to generalize poorly to previously unseen fabrics, posing a crucial limitation to their applicability. Moreover, current research focuses on adapting converged models to previously unseen fabrics in a post hoc manner, rather than training models that generalize better in the first place. In our work, we explore this potential for the first time. Specifically, we propose that previously unseen fabrics can be regarded as shifts in the underlying data distribution. We therefore argue that factors which reportedly improve a model’s resistance to distribution shifts should also improve the performance of supervised fabric AD methods on unseen fabrics. Hence, we assess the potential benefits of: (I) vicinal risk minimization (VRM) techniques adapted to the fabric AD use-case, (II) different loss functions, (III) ImageNet pre-training, (IV) dataset diversity, and (V) model architecture as well as model complexity. The subsequently performed large-scale analysis reveals that (I) only the VRM technique, AugMix, consistently improves performance on unseen fabrics; (II) hypersphere classifier outperforms other loss functions when combined with AugMix and (III) ImageNet pre-training, which is already beneficial on its own; (IV) increasing dataset diversity improves performance on unseen fabrics; and (V) architectures with better ImageNet performance also perform better on unseen fabrics, yet the same does not hold for more complex models. Notably, the results show that not all factors and techniques which reportedly improve a model’s resistance to distribution shifts in natural images also improve the generalization of supervised fabric AD methods to unseen fabrics, demonstrating the necessity of our work. Additionally, we also assess whether the performance gains of models which generalize better propagate to post hoc adaptation methods and show this to be the case. Since no suitable fabric dataset was publicly available at the time of this work, we acquired our own fabric dataset, called OLP, as the basis for the above experiments. OLP consists of 38 complex, patterned fabrics, more than 6400 images in total, and is made publicly available.Keywords
Funding Information
- German Federation of Industrial Research Associations (AiF) (grant number 19811 N, FKZ: 01IS21049C)
- German Federal Ministry of Education and Research (BMBF) (grant number 19811 N, FKZ: 01IS21049C)
This publication has 48 references indexed in Scilit:
- Fabric defect detection systems and methods—A systematic literature reviewOptik, 2016
- Deep Residual Learning for Image Recognition2016
- Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification2015
- Anomaly detectionACM Computing Surveys, 2009
- The Elements of Statistical Learning2009
- The relationship between Precision-Recall and ROC curves2006
- Defect detection and classification on web textile fabric using multiresolution decomposition and neural networks2003
- A Survey of Automated Visual InspectionComputer Vision and Image Understanding, 1995
- FDAS: A Knowledge-based Framework for Analysis of Defects in Woven Textile StructuresThe Journal of the Textile Institute, 1992
- Estimating the Dimension of a ModelThe Annals of Statistics, 1978