AI-supported quality assurance for image analysis

Every hour your experts spend on manual image analysis or report generation is time lost elsewhere. We take on precisely this burden: We develop AI systems that automatically classify, segment, and measure microstructures, fracture surfaces, micrographs, and CT data—reproducibly across laboratories and shifts. We train AI models using your own image data, validate them against technical references, and build reporting assistants that consolidate measurement data, images, and findings into drafts ready for review. Technical approval remains with your engineers—we provide the tool. The result: less effort in data analysis, faster batch approvals and material qualifications, and lower scrap and warranty risks—all without having to hire additional staff.

Automated image and microstructure analysis

We develop AI-based visual analysis systems for the automatic classification, segmentation, and measurement of features in micrographs, SEM/EBSD images, CT volumes, and fracture surfaces, tailored to your specific image data. This reduces the need for manual evaluation, ensures reproducible results across laboratories and shifts, accelerates the approval of new batches and material qualifications, and lowers your risk of scrap and warranty claims.

R&D services:

• Automatic classification and segmentation of grain size, phase fractions, and pores
• Detection and measurement of inclusions, cracks, and damage patterns
• 3D reconstruction of features from CT volumes
• Customization of the analysis system to your specific image data and testing procedures
• Validation of the hit rate and measurement uncertainty using technical reference cases

AI for non-compliant image data

We develop hybrid deep learning and rule-based image processing methods for standards-compliant microstructure evaluations, ranging from data-sparse training methods to foundation models designed to reduce training data requirements. This lowers the cost of microstructure evaluation and delivers comparable, reliable results across companies.

R&D services:

• Combination of deep learning and rule-based methods for standard-compliant evaluations
• Use of foundation models to reduce training data requirements
• Processing of heterogeneous image qualities and specimen preparation types
• Alignment of assessments with relevant testing standards
• Establishment of comparable, reliable evaluation criteria across locations

AI report assistant for inspection and damage analyses

We are developing an AI assistant that generates a structured report draft in your custom format based on measurement data, inspection images, and evaluations from a current inspection order—including automatically labeled images and source links. Final approval remains with the responsible engineer. This allows you to reduce report generation time, ensure consistent quality across engineers and shifts, and increase throughput without additional staff.

R&D sercives:

• Automatic consolidation of measurement data, images, and evaluations into a draft report
• Automatic labeling of figures and linking to source data
• Customization of the wizard to your specific report format
• Setup of an approval workflow for responsible engineers
• Integration of the wizard into existing LIMS or QMS environments

to top