Technical cleanliness: AI helps to detect particles
In order to optimise the classification of process-critical particles in the technical cleanliness process, Zeiss has been offering the Zen Core Technical Cleanliness Analysis module with pre-trained machine learning-based object classification models for its microscopes since January 2022. This module automates the type classification of particles based on a classic grey value determination. For this purpose, the results obtained from the classical analysis are combined to form a large number of uncorrelated decision trees, including size, shape, intensity and type classification.
Module was developed with correctly classified particles
This module applies these classifications, developed by Zeiss or individually trained by the user, to these decision trees. If particles are detected that were previously incorrectly classified as non-metallic particles, the machine learning-based object classification overwrites the results of the classic grey value determination.
Better comparability of technical cleanliness results
The use of the new object classification pays off in several ways: The results are not only more discriminating without a manual test. Companies that use the module also reduce the workload of their operators and increase the comparability of the results. In addition, the microscopy solutions are no longer blocked by the otherwise necessary rework, which means that the utilisation of the devices increases.
The above texts, or parts thereof, were automatically translated from the original language text using a translation system (DeepL API).
Despite careful machine processing, translation errors cannot be ruled out.