Automated workflow based on identification of calibration samples

Abstract

Automated workflow based on identification of calibration samples. A method for calibrating a microscope system and a corresponding calibration system are presented. The method includes recording an overview image of a sample stage of the microscope system and identifying a calibration sample in the recorded overview image. Furthermore, the method comprises classifying the calibration sample in the recorded overview image into one of a plurality of calibration sample classes by means of a classification system which is trained using the training data in order to form a model such that the classification system is adapted Used to classify unknown input data into predicted classes. Furthermore, the method comprises selecting a calibration workflow for calibrating the microscope system based on the sorted calibration sample class, wherein the selection is performed by means of a workflow indicator value used as input for the workflow selection system value. Additionally, the method may include calibrating components of the microscope system.

Description

technical field [0001] The invention relates to the calibration of microscope systems, and in particular to a computer-implemented method for calibrating a microscope system, a corresponding calibration system and a corresponding computer program product. Background technique [0002] Microscope systems are not only used for imaging, but are increasingly also used for measurement purposes. In addition, data from imaging systems are increasingly being used to obtain statistical data (larger number of measurements to verify results with facts / statistics, or QM for part manufacturing, etc.), which also requires the use of various system (parallelization). In addition, in the case of objectives for example, there are manufacturing ranges (NA and magnification, different CORR ring settings) which provide different results when using different microscopes. [0003] For this reason, the optical system has to be calibrated, again to counteract variations eg from manufacturing or r...

AI Technical Summary

Problems solved by technology

[0006] Previous calibration methods often have several disadvantages: Usually, calibration has to be started manually and explicitly; additionally, calibration samples have to be selected by hand or with the help of a manual

Furthermore, there may be a lack of reproducibility of the calibration measures, since the user performs them entirely by hand

[0008] Also, it should be considered that servicing the microscope system requires an educated user and it may take a long time for someone to acquire the necessary knowledge

Furthermore, in the case of multiple system installations, a considerable expenditure of time and personnel is required, since service other than limited service / assistance via remote service is often not possible

Images