Bitte benutzen Sie diese Referenz, um auf diese Ressource zu verweisen:
doi:10.22028/D291-40382
Titel: | Reproducible Quantification of the Microstructure of Complex Quenched and Quenched and Tempered Steels Using Modern Methods of Machine Learning |
VerfasserIn: | Bachmann, Björn-Ivo Müller, Martin Britz, Dominik Staudt, Thorsten Mücklich, Frank |
Sprache: | Englisch |
Titel: | Metals |
Bandnummer: | 13 |
Heft: | 8 |
Verlag/Plattform: | MDPI |
Erscheinungsjahr: | 2023 |
Freie Schlagwörter: | microstructure classification microstructure segmentation machine learning quenched steel martensite bainite |
DDC-Sachgruppe: | 500 Naturwissenschaften |
Dokumenttyp: | Journalartikel / Zeitschriftenartikel |
Abstract: | Current conventional methods of evaluating microstructures are characterized by a high degree of subjectivity and a lack of reproducibility. Modern machine learning (ML) approaches have already shown great potential in overcoming these challenges. Once trained with representative data in combination with objective ground truth, the ML model is able to perform a task properly in a reproducible and automated manner. However, in highly complex use cases, it is often not possible to create a definite ground truth. This study addresses this problem using the underlying showcase of microstructures of highly complex quenched and quenched and tempered (Q/QT) steels. A patch-wise classification approach combined with a sliding window technique provides a solution for segmenting entire microphotographs where pixel-wise segmentation is not applicable since it is hardly feasible to create reproducible training masks. Using correlative microscopy, consisting of light optical microscope (LOM) and scanning electron microscope (SEM) micrographs, as well as corresponding data from electron backscatter diffraction (EBSD), a training dataset of reference states that covers a wide range of microstructures was acquired in order to train accurate and robust ML models in order to classify LOM or SEM images. Despite the enormous complexity associated with the steels treated here, classification accuracies of 88.8% in the case of LOM images and 93.7% for high-resolution SEM images were achieved. These high accuracies are close to super-human performance, especially in consideration of the reproducibility of the automated ML approaches compared to conventional methods based on subjective evaluations through experts. |
DOI der Erstveröffentlichung: | 10.3390/met13081395 |
URL der Erstveröffentlichung: | https://doi.org/10.3390/met13081395 |
Link zu diesem Datensatz: | urn:nbn:de:bsz:291--ds-403821 hdl:20.500.11880/36304 http://dx.doi.org/10.22028/D291-40382 |
ISSN: | 2075-4701 |
Datum des Eintrags: | 28-Aug-2023 |
Fakultät: | NT - Naturwissenschaftlich- Technische Fakultät |
Fachrichtung: | NT - Materialwissenschaft und Werkstofftechnik |
Professur: | NT - Prof. Dr. Frank Mücklich |
Sammlung: | SciDok - Der Wissenschaftsserver der Universität des Saarlandes |
Dateien zu diesem Datensatz:
Datei | Beschreibung | Größe | Format | |
---|---|---|---|---|
metals-13-01395.pdf | 14,72 MB | Adobe PDF | Öffnen/Anzeigen |
Diese Ressource wurde unter folgender Copyright-Bestimmung veröffentlicht: Lizenz von Creative Commons