High-speed robot vision and surface defect classification of 3D-shaped products and components
In Europe, sales in the billions of euros are associated with customers being given a flawless impression of the quality of products. In many fields, the quality inspection of product surfaces is therefore an integral part of quality management. An increasing number of complex shaped products and components cannot be inspected manually in the short time of the production cycle. This problem can only be solved by automated inspection systems that can detect, classify and evaluate defects according to product-specific defect catalogs. At the PCCL we have developed inspection robots that make full surface inspections of complex 3D-parts in less than 10 seconds. By means of artificial intelligence, the system is able to recognize and correctly classify numerous defect types. That’s not all. As a special feature, it can precisely distinguish permitted anomalies from non-permitted defects and thus considerably increase the yield of good parts.
Dieter P. Gruber studied Technical Physics at the Institute for Theoretical Physics at Graz University of Technology. Due to his combined specialization in information technology and material physics he moved to the Fraunhofer Institute for Solar Energy Systems (ISE) in Freiburg during his phD-theses in 2001-2002. In2003 Dieter P. Gruber started his research work at the Polymer Competence Center Leoben. Dieter P. Gruber is responsible for the development of fundamentally new methods for surface characterization and robotic vision by the use of machine- and deep-learning. He habilitated to assoc. Prof. in 2015 at the University of Leoben and he is the author of more than 70 publications and 12 patents. He awarded numerous research prizes such as the Magna ACS Innovation Award, the Houska Prize and the SFG Fast Forward Award. He was Austrian of the Year 2014 in the category “Research”.
