Maria del Mar Juez Lorenzo, Vladislav Kolarik, Khyati Sethia, Petr Strakos
{"title":"Segmentation and Metallographic Evaluation of Aluminium Slurry Coatings Using Machine Learning Techniques","authors":"Maria del Mar Juez Lorenzo, Vladislav Kolarik, Khyati Sethia, Petr Strakos","doi":"10.1007/s11085-024-10321-3","DOIUrl":null,"url":null,"abstract":"<div><p>Analysis of scanning electron microscope (SEM) images is crucial for characterising aluminide diffusion coatings deposited via the slurry route on steels, yet challenging due to various factors like imaging artefacts, noise, and overlapping features such as resin, precipitates, cracks, and pores. This study focuses on determining the thicknesses of the coating layers Fe<sub>2</sub>Al<sub>5</sub> and, if present, FeAl, pore characteristics, and chromium precipitate fractions after the heat treatment that forms the diffusion coating. A deep learning SEM image segmentation model utilising U-Net architecture is proposed. Ground truth data were generated using the trainable Weka segmentation plugin in ImageJ, manually refined for accuracy, and supplemented with synthetic data from Blender 3D software for data augmentation of a limited number of SEM label images. The deep learning model trained on a combination of synthetic and real SEM data achieved mean dice scores of 98.7% ± 0.2 for the Fe<sub>2</sub>Al<sub>5</sub> layer, 82.6% ± 8.1 for pores, and 81.48% ± 3.6 for precipitates when evaluated on manually labelled SEM data. The deep learning procedure was applied to evaluate a series of SEM images of diffusion coatings obtained with three different slurry compositions. The evaluation revealed that using a slurry without a rheology modifier may lead to a thicker partial Fe<sub>2</sub>Al<sub>5</sub> layer that is formed by inward diffusion. The relation between the outward and inward diffusion Fe<sub>2</sub>Al<sub>5</sub> layers was not affected by the coating thickness. The thinner diffusion coating presents lower pores and chromium precipitate fractions independently of the slurry selected.</p></div>","PeriodicalId":724,"journal":{"name":"Oxidation of Metals","volume":"101 6","pages":"1497 - 1512"},"PeriodicalIF":2.1000,"publicationDate":"2024-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s11085-024-10321-3.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Oxidation of Metals","FirstCategoryId":"88","ListUrlMain":"https://link.springer.com/article/10.1007/s11085-024-10321-3","RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"METALLURGY & METALLURGICAL ENGINEERING","Score":null,"Total":0}
引用次数: 0
Abstract
Analysis of scanning electron microscope (SEM) images is crucial for characterising aluminide diffusion coatings deposited via the slurry route on steels, yet challenging due to various factors like imaging artefacts, noise, and overlapping features such as resin, precipitates, cracks, and pores. This study focuses on determining the thicknesses of the coating layers Fe2Al5 and, if present, FeAl, pore characteristics, and chromium precipitate fractions after the heat treatment that forms the diffusion coating. A deep learning SEM image segmentation model utilising U-Net architecture is proposed. Ground truth data were generated using the trainable Weka segmentation plugin in ImageJ, manually refined for accuracy, and supplemented with synthetic data from Blender 3D software for data augmentation of a limited number of SEM label images. The deep learning model trained on a combination of synthetic and real SEM data achieved mean dice scores of 98.7% ± 0.2 for the Fe2Al5 layer, 82.6% ± 8.1 for pores, and 81.48% ± 3.6 for precipitates when evaluated on manually labelled SEM data. The deep learning procedure was applied to evaluate a series of SEM images of diffusion coatings obtained with three different slurry compositions. The evaluation revealed that using a slurry without a rheology modifier may lead to a thicker partial Fe2Al5 layer that is formed by inward diffusion. The relation between the outward and inward diffusion Fe2Al5 layers was not affected by the coating thickness. The thinner diffusion coating presents lower pores and chromium precipitate fractions independently of the slurry selected.
期刊介绍:
Oxidation of Metals is the premier source for the rapid dissemination of current research on all aspects of the science of gas-solid reactions at temperatures greater than about 400˚C, with primary focus on the high-temperature corrosion of bulk and coated systems. This authoritative bi-monthly publishes original scientific papers on kinetics, mechanisms, studies of scales from structural and morphological viewpoints, transport properties in scales, phase-boundary reactions, and much more. Articles may discuss both theoretical and experimental work related to gas-solid reactions at the surface or near-surface of a material exposed to elevated temperatures, including reactions with oxygen, nitrogen, sulfur, carbon and halogens. In addition, Oxidation of Metals publishes the results of frontier research concerned with deposit-induced attack. Review papers and short technical notes are encouraged.