State Secretary Dr. Denis Alt Hands Over Funding Decision in Kaiserslautern

Project Equality: Finding Defects in Interfaces Faster Thanks to Artificial Intelligence

Press Release / January 15, 2024

From household items to vehicle components and medical technology – when manufacturing products, even the smallest defects need to be detected reliably and quickly, preferably automatically. In the project »eQuality«, researchers at the Fraunhofer Institute for Industrial Mathematics ITWM are investigating how this process can be improved thanks to artificial intelligence (AI). The state of Rhineland-Palatinate is supporting the project with 600,000 euros.

Although automated inspection systems are state of the art, they are expensive and require a lot of time and expertise to develop. The latter applies in particular to troubleshooting in non-standardized components. A well-stocked database that maps application-relevant defects is faster. This not only collects real image data, but also data records that describe defects in such a way that artificial and photorealistic images can be generated from them.

Digital Defect Library

The department »Image Processing« at the Fraunhofer ITWM in Kaiserslautern has long been involved in the detection of defects in surfaces: »We supply complete inspection systems, i.e. hardware and software,« says department head Markus Rauhut. »In recent years, we have expanded our algorithms to include machine learning methods. We can now generate photorealistic data sets that we use to train our AI.« Project manager Dr. Petra Gospodnetic adds: »This allows us to significantly reduce the number of physical error samples from which the AI usually learns. Thanks to the funding from the state of Rhineland-Palatinate, we can significantly expand our error library and recognize and specify even more errors.«

© Fraunhofer ITWM
State Secretary Dr. Denis Alt, ITWM Director Prof. Dr. Anita Schöbel and Project Manager Dr. Petra Gospodnetic (from left to right)

Dr. Denis Alt, State Secretary in the Ministry of Science and Health, personally brought the funding decision to Kaiserslautern and classified eQuality as follows: »The project shows that artificial intelligence already has an enormous application proximity and can contribute to making production more sustainable and ensuring the quality of products. With eQuality, the Fraunhofer ITWM is helping to strengthen the innovative power of the research location and the competitiveness and future viability of our economy by contributing to ensuring Rhineland-Palatinate's pioneering role in the field of AI, a declared goal in our coalition agreement.«

Initial Focus on Defects on Metal Surfaces

The aim of the project »eQuality« is to create an online platform that people from industry and research can use to generate and download defects corresponding to their specific products. As a first step, the library covers a range of defects that are characteristic of metal surfaces – such as dents, scratches, cracks, coolant residues, metal chips, dust particles or stains.