Hannover Messe 2022: Intelligent inspection system categorises workpieces and their properties based on images

The use of artificial intelligence (AI) methods is on the rise. Even in industry, they play an essential role. However, small and medium-sized enterprises (SMEs) tend to be critical of this technology. Researchers at Technische Universität Kaiserslautern (TUK) seek to change their scepticism and want to make AI processes transparent and explainable to SMEs. At the Hannover Messe at the Rhineland-Palatinate research stand (Hall 2, Stand B40) from 30 May to 3 June, they will be presenting an image-based inspection system the team has developed. The system detects workpieces based on specific criteria. Therefore, the system uses machine learning methods, an AI process, and classifies the components accordingly.

AI technologies allow many processes in the industry to run quickly and efficiently, for example, in quality inspection through image recognition or in robot-controlled work in manufacturing. But many SMEs, in particular, shy away from investing in them. “In many cases, one buys ready-made software, a black box, so to speak, of which one does not know exactly what is inside”, explains Patrick Rüdiger-Flore from the Institute for Manufacturing Technology and Production Systems at TU Kaiserslautern.

The Institute's team has set itself the task of making AI systems more explainable and transparent for SMEs. This is also true for a technology presented at the trade fair. The team has developed an image-based inspection system to recognise and classify workpieces. “Our software uses machine learning, an artificial intelligence technique,” explains the engineer. “The system learns based on the data presented to it.”

The technology behind it is relatively simple: The workpiece is placed on a control station. A camera photographs the component and automatically sends the images to a computer, where software evaluates the image. “The system uses certain criteria to recognise the workpieces and variations of the workpieces.”

To show interested companies at the trade fair which features the technology uses, the images contain markings with which AI determines precisely which workpiece is involved. “You can compare this well with your intuition,” says Partick Rüdiger-Flore. “Our method compares the markings with images in a database, to which new images are also added. In this way, the system also learns.”
The project is supported by the European Union (EU) through the Regional Development Fund and the Commercial Vehicle Cluster (CVC) Southwest.
Questions can be directed to:
Dr-Ing. Patrick Rüdiger-Flore
Institute for Manufacturing Technology and Production Systems
Phone: +49 631 205-4282
E-mail: patrick.ruediger(at)mv.uni-kl.de

+++
Klaus Dosch, Department of Technology and Innovation, is organizing the presentation of the researchers of the TU Kaiserslautern at the fair. He is the contact partner for companies and, among other things, establishes contacts to science.
Contact: Klaus Dosch, E-mail: dosch[at]rti.uni-kl.de, Phone: +49 631 205-3001