Long, Emily R;
Bone, Alayna;
Breitung, Eric M;
Thickett, David;
Grau-Bové, Josep;
(2022)
Automated corrosion detection in Oddy test coupons using convolutional neural networks.
Heritage Science
, 10
(1)
, Article 150. 10.1186/s40494-022-00778-3.
Preview |
Text
s40494-022-00778-3.pdf - Published Version Download (2MB) | Preview |
Abstract
The Oddy test is an accelerated ageing test used to determine whether a material is appropriate for the storage, transport, or display of museum objects. The levels of corrosion seen on coupons of silver, copper, and lead indicate the material’s safety for use. Although the Oddy test is conducted in heritage institutions around the world, it is often critiqued for a lack of repeatability. Determining the level of corrosion is a manual and subjective process, in which outcomes are affected by differences in individuals’ perceptions and practices. This paper proposes that a more objective evaluation can be obtained by utilising a convolutional neural network (CNN) to locate the metal coupons and classify their corrosion levels. Images provided by the Metropolitan Museum of Art (the Met) were labelled for object detection and used to train a CNN. The CNN correctly identified the metal type and corrosion level of 98% of the coupons in a test set of the Met’s images. Images were also collected from the American Institute for Conservation’s Oddy test wiki page. These images suffered from low image quality and were missing the classification information needed to train the CNN. Experts from cultural heritage institutions evaluated the coupons in the images, but there was a high level of disagreement between expert classifications. Therefore, these images were not used to train the CNN. However, the images proved useful in testing the limitations of the CNN trained on the Met’s data when applied to images of coupons from different Oddy test protocols and photo documentation procedures. This paper presents the effectiveness of the CNN trained on the Met’s data to classify Met and non-Met Oddy test coupons. Finally, this paper proposes the next steps needed to produce a universal CNN-based classification tool. Graphic Abstract
| Type: | Article |
|---|---|
| Title: | Automated corrosion detection in Oddy test coupons using convolutional neural networks |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1186/s40494-022-00778-3 |
| Publisher version: | https://doi.org/10.1186/s40494-022-00778-3 |
| Language: | English |
| Additional information: | © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. |
| Keywords: | Object detection, Machine learning, Computer vision, Preventive conservation |
| UCL classification: | UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment > Bartlett School Env, Energy and Resources UCL > Provost and Vice Provost Offices > UCL BEAMS UCL |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10157385 |
Archive Staff Only
 |
View Item |
