Machine learning and process digitization

Topic

© Fraunhofer IKTS
Process digitization using deep learning-based code localization and subsequent image segmentation (red overlay) for qualitative evaluation of the printed image.

The working group "Optical Test Methods and Nanosensors" focuses primarily on the development of novel sensor materials and their application to generate data in industrial production processes. With the use of machine learning and comprehensive data analysis, the added value of the IKTS and the customer offer are expanded. Data science methods are used in the development of new materials, for the analysis and optimization of sensor signals and ultimately for the optimization of production processes. Thus, customers can generate more accurate measurement data, integrate sensors into processes in a fail-safe manner, and implement new applications, such as condition monitoring and predictive maintenance.

 

Services offered

 

  • Digitization of production processes with the aid of optical inspection methods and machine learning
  • Concepts for condition monitoring of processes and predictive maintenance