Diffusion-Based Image Generation for In-Distribution Data Augmentation in Surface Defect Detection

Luigi Capogrosso, Federico Girella, Francesco Taioli, Michele Chiara, Muhammad Aqeel, Franco Fummi, Francesco Setti, Marco Cristani

2024

Abstract

In this study, we show that diffusion models can be used in industrial scenarios to improve the data augmentation procedure in the context of surface defect detection. In general, defect detection classifiers are trained on ground-truth data formed by normal samples (negative data) and samples with defects (positive data), where the latter are consistently fewer than normal samples. For these reasons, state-of-the-art data augmentation procedures add synthetic defect data by superimposing artifacts to normal samples. This leads to out-of-distribution augmented data so that the classification system learns what is not a normal sample but does not know what a defect really is. We show that diffusion models overcome this situation, providing more realistic in-distribution defects so that the model can learn the defect’s genuine appearance. We propose a novel approach for data augmentation that mixes out-of-distribution with in-distribution samples, which we call In&Out. The approach can deal with two data augmentation setups: i) when no defects are available (zero-shot data augmentation) and ii) when defects are available, which can be in a small number (few-shot) or a large one (full-shot). We focus the experimental part on the most challenging benchmark in the state-of-the-art, i.e., the Kolektor Surface-Defect Dataset 2, defining the new state-of-the-art classification AP score under weak supervision of .782. The code is available at https://github.com/intelligolabs/in and out.

Paper Citation

in Harvard Style

Capogrosso L., Girella F., Taioli F., Chiara M., Aqeel M., Fummi F., Setti F. and Cristani M. (2024). Diffusion-Based Image Generation for In-Distribution Data Augmentation in Surface Defect Detection. In Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP; ISBN 978-989-758-679-8, SciTePress, pages 409-416. DOI: 10.5220/0012350400003660

in Bibtex Style

@conference{visapp24,
author={Luigi Capogrosso and Federico Girella and Francesco Taioli and Michele Chiara and Muhammad Aqeel and Franco Fummi and Francesco Setti and Marco Cristani},
title={Diffusion-Based Image Generation for In-Distribution Data Augmentation in Surface Defect Detection},
booktitle={Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP},
year={2024},
pages={409-416},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012350400003660},
isbn={978-989-758-679-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 19th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 2: VISAPP
TI - Diffusion-Based Image Generation for In-Distribution Data Augmentation in Surface Defect Detection
SN - 978-989-758-679-8
AU - Capogrosso L.
AU - Girella F.
AU - Taioli F.
AU - Chiara M.
AU - Aqeel M.
AU - Fummi F.
AU - Setti F.
AU - Cristani M.
PY - 2024
SP - 409
EP - 416
DO - 10.5220/0012350400003660
PB - SciTePress