Mastering complexity. ​With Microscopy Solutions from ZEISS.​

The tasks in materialography are constantly changing and the performance requirements are increasing. Components are becoming ever more complex, coatings ever thinner. Miniaturization in electronics is progressing. New composite materials are being used in lightweight construction in particular. Sustainable materials and additively manufactured components pose particular challenges in terms of material characterization. At the same time the defect tolerances are generally decreasing, making more precise analyses are required. Materials testing laboratories must keep pace with these developments. They are under increasing time pressure. ZEISS offers materialographers a unique depth of microscopes, software and services to effectively solve the testing effectively. In addition, ZEISS Microscopy Solutions make it particularly easy for all users, because thanks to deep learning algorithms, results are more reliable and faster than ever before.

Faster and more reliable ­testing using deep learning

ZEISS ZEN core provides automated image analysis for particle and phase analysis. Deep learning techniques significantly reduce the workload for these applications and always work reproducibly, according to the same trained patterns, regardless of the user and location. The microscope image of an aluminium alloy is automatically segmented using deep learning. The algorithm reliably identifies individual grains. Touching and overlapping objects can be separated (instance segmentation).

Segmentation from ZEISS arivis Cloud

Latest deep learning models: Automatic AI-supported determination of the grain boundaries of a ceramic, an austenitic steel, a hard metal and a copper-based alloy. In this example, it was sufficient to train a deep learning model to reliably recognize the grain boundaries of all four materials.

Correlation with light, digital and electron microscopes

ZEISS ZEN core tool kits