Grain size determination: Best practice in the working group
The first topic was the determination of grain sizes. Mr. Bachmann from Saarland University presented approaches to grain size determination using machine learning. Correlative microscopy and the development of the automated etching microscope "ThEtching" can be used to automatically determine the structural components of steels. Dr.-Ing. Korpala from MiViA GmbH presented a precise AI-based method for analyzing steel microstructures.
The topic of grain size determination was rounded off by the contribution of Dr. Witte from Salzgitter Mannesmann Forschung. He presented the results of the tests carried out in the working group to determine grain boundaries. A total of 27 participants from 12 laboratories took part. The results were used to develop approaches for best practice. The working group will continue to deal with this topic in future experiments.
AI methods and other topics
The second main topic, AI methods, was also vividly illustrated by exciting presentations. Mr. Goedecke from GFaI presented the DIAgraph ML project for the classification of graphite structures in cast iron using machine learning methods. Mr. Jansche from Aalen University gave an overview of the use of machine learning in material microscopy. Further topics were discussed, including insights into magnification levels and 3D metallography systems.
Enriching exchange
After the exciting presentations, the participants gladly took the opportunity for further exchange and networking. The "Quantitative Microstructural Analysis" working group would like to thank Mr. Kaip and the GFaI in Berlin for their hospitality and successful organization.
The next meeting is planned for spring 2025 in Freiberg at the MiViA company and the Institute of Metal Forming.