Version 1
: Received: 18 October 2024 / Approved: 21 October 2024 / Online: 21 October 2024 (12:06:01 CEST)
How to cite:
Ugwu, C. I.; Caruso, E.; Lanz, O. Fractals as Pre-training Datasets for Anomaly Detection and Localization. Preprints2024, 2024101579. https://doi.org/10.20944/preprints202410.1579.v1
Ugwu, C. I.; Caruso, E.; Lanz, O. Fractals as Pre-training Datasets for Anomaly Detection and Localization. Preprints 2024, 2024101579. https://doi.org/10.20944/preprints202410.1579.v1
Ugwu, C. I.; Caruso, E.; Lanz, O. Fractals as Pre-training Datasets for Anomaly Detection and Localization. Preprints2024, 2024101579. https://doi.org/10.20944/preprints202410.1579.v1
APA Style
Ugwu, C. I., Caruso, E., & Lanz, O. (2024). Fractals as Pre-training Datasets for Anomaly Detection and Localization. Preprints. https://doi.org/10.20944/preprints202410.1579.v1
Chicago/Turabian Style
Ugwu, C. I., Emanuele Caruso and Oswald Lanz. 2024 "Fractals as Pre-training Datasets for Anomaly Detection and Localization" Preprints. https://doi.org/10.20944/preprints202410.1579.v1
Abstract
Anomaly detection is crucial in large-scale industrial manufacturing as it helps detect and localise defective parts. Pre-training feature extractors on large-scale datasets is a popular approach for this task. Stringent data security, privacy regulations, high costs and acquisition time hinder the development of large-scale datasets for training and benchmarking. Despite recent work focusing primarily on the development of new anomaly detection methods based on such extractors, not much attention has been paid to the importance of the data used for pre-training. This study compares representative models pre-trained with fractal images against those pre-trained with ImageNet, without subsequent task-specific fine-tuning. We evaluated the performance of eleven state-of-the-art methods on MVTecAD, MVTec LOCO AD, and VisA, well-known benchmark datasets inspired by real-world industrial inspection scenarios. Further, we propose a novel method to create a dataset by combining the dynamically generated fractal images creating a ``Multi-Formula'' dataset. Even though pre-training with ImageNet leads to better results, fractals can achieve close performance to ImageNet under proper parametrisation. This opens up the possibility for a new research direction where feature extractors could be trained on synthetically generated abstract datasets mitigating the ever-increasing demand for data in machine learning while circumventing privacy and security concerns.
Keywords
Fractals; Mandelbulb; Anomaly detection; Industrial inspection; Synthetic data
Subject
Computer Science and Mathematics, Artificial Intelligence and Machine Learning
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
