Marie Stiefel, Martin Müller, Björn-Ivo Bachmann, Maria Agustina Guitar, Ullal Pranav Nayak, Frank Mücklich
{"title":"加强微结构的机器学习分类:在 CNN 中加入图像数据和元数据的工作流程研究","authors":"Marie Stiefel, Martin Müller, Björn-Ivo Bachmann, Maria Agustina Guitar, Ullal Pranav Nayak, Frank Mücklich","doi":"10.1557/s43579-024-00549-0","DOIUrl":null,"url":null,"abstract":"<p>In view of the paradigm shift toward data-driven research in materials science and engineering, handling large amounts of data becomes increasingly important. The application of FAIR (findable, accessible, interoperable, reusable) data principles emphasizes the importance of metadata describing datasets. We propose a novel data processing and machine learning (ML) pipeline to extract metadata from micrograph image files, then combine image data and their metadata for microstructure classification with a deep learning approach compared to a classic ML approach. The ML model attained excellent performances with and without metadata and bears potential for performance improvement of further use cases within the community.</p><h3 data-test=\"abstract-sub-heading\">Graphical abstract</h3>\n","PeriodicalId":19016,"journal":{"name":"MRS Communications","volume":null,"pages":null},"PeriodicalIF":1.8000,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Enhancing machine learning classification of microstructures: A workflow study on joining image data and metadata in CNN\",\"authors\":\"Marie Stiefel, Martin Müller, Björn-Ivo Bachmann, Maria Agustina Guitar, Ullal Pranav Nayak, Frank Mücklich\",\"doi\":\"10.1557/s43579-024-00549-0\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>In view of the paradigm shift toward data-driven research in materials science and engineering, handling large amounts of data becomes increasingly important. The application of FAIR (findable, accessible, interoperable, reusable) data principles emphasizes the importance of metadata describing datasets. We propose a novel data processing and machine learning (ML) pipeline to extract metadata from micrograph image files, then combine image data and their metadata for microstructure classification with a deep learning approach compared to a classic ML approach. The ML model attained excellent performances with and without metadata and bears potential for performance improvement of further use cases within the community.</p><h3 data-test=\\\"abstract-sub-heading\\\">Graphical abstract</h3>\\n\",\"PeriodicalId\":19016,\"journal\":{\"name\":\"MRS Communications\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.8000,\"publicationDate\":\"2024-04-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"MRS Communications\",\"FirstCategoryId\":\"88\",\"ListUrlMain\":\"https://doi.org/10.1557/s43579-024-00549-0\",\"RegionNum\":4,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"MATERIALS SCIENCE, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"MRS Communications","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.1557/s43579-024-00549-0","RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"MATERIALS SCIENCE, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
摘要
鉴于材料科学与工程领域的研究模式正在向数据驱动型转变,处理大量数据变得越来越重要。FAIR(可查找、可访问、可互操作、可重用)数据原则的应用强调了描述数据集的元数据的重要性。我们提出了一种新颖的数据处理和机器学习(ML)管道,用于从显微摄影图像文件中提取元数据,然后与传统的 ML 方法相比,利用深度学习方法将图像数据及其元数据结合起来,用于显微结构分类。该 ML 模型在有元数据和无元数据的情况下均表现出色,具有在社区内进一步使用案例中提高性能的潜力。
Enhancing machine learning classification of microstructures: A workflow study on joining image data and metadata in CNN
In view of the paradigm shift toward data-driven research in materials science and engineering, handling large amounts of data becomes increasingly important. The application of FAIR (findable, accessible, interoperable, reusable) data principles emphasizes the importance of metadata describing datasets. We propose a novel data processing and machine learning (ML) pipeline to extract metadata from micrograph image files, then combine image data and their metadata for microstructure classification with a deep learning approach compared to a classic ML approach. The ML model attained excellent performances with and without metadata and bears potential for performance improvement of further use cases within the community.
期刊介绍:
MRS Communications is a full-color, high-impact journal focused on rapid publication of completed research with broad appeal to the materials community. MRS Communications offers a rapid but rigorous peer-review process and time to publication. Leveraging its access to the far-reaching technical expertise of MRS members and leading materials researchers from around the world, the journal boasts an experienced and highly respected board of principal editors and reviewers.