Abstract
Transparent materials are employed for creating different kinds of products and have to meet high quality requirements. First of all, transparent materials have to be free from scattering defects, e.g., enclosed air bubbles. Visual inspection systems based on dark field setups are principally capable of imaging these kinds of defects, however, it usually requires much effort to adapt them to the test object on hand. This article shows how light transport matrices can be calculated for an optical system consisting of a programmable area light source and a telecentric camera. Two features are proposed that can be extracted out of these matrices and that allow to image scattering defects present in a transparent object without the need of adapting the system to the actual test object. The results of synthetic experiments obtained using a physically based and adequately extended rendering framework approved the proposed approach and showed that it even outperforms classical inspection systems in some situations.
Zusammenfassung
Transparente Materialien werden zur Herstellung verschiedenster Produkte verwendet und müssen hohen Qualitätsanforderungen genügen. Insbesondere müssen sie frei sein von streuenden Defekten wie beispielsweise eingeschlossenen Luftblasen. Prinzipiell sind auf Dunkelfeldansätzen basierende Inspektionssysteme dazu in der Lage, diese Defekte abzubilden. Allerdings muss die geometrische Ausrichtung der einzelnen Systemkomponenten häufig an das vorliegende Prüfobjekt angepasst werden, was mit einem hohen manuellen Einrichtaufwand einhergeht. Dieser Beitrag zeigt, wie für ein optisches System bestehend aus einer telezentrischen Kamera und einer programmierbaren Flächenlichtquelle sogenannte Lichttransportmatrizen berechnet werden können. Es werden zwei Merkmale vorgestellt, die sich aus diesen Matrizen extrahieren lassen und die eine Abbildung von in transparenten Prüfobjekten enthaltenen streuenden Defekten ermöglichen, ohne dass das System an den Prüfling angepasst werden muss. Synthetische Experimente, die mit Hilfe eines entsprechend erweiterten physikalisch basierten Renderingframeworks durchgeführt wurden, bestätigten die Eignung des vorgeschlagenen Verfahrens. Zudem wird gezeigt, dass der Ansatz herkömmlichen Inspektionsverfahren in einigen Situationen sogar überlegen ist.
About the authors
In 2012 Johannes Meyer received his B.Sc. and in 2014 his M.Sc. in computer science from the Karlsruhe Institute of Technology (KIT) in Germany. Since then he is a PhD student at the Vision and Fusion Laboratory (IES) at the KIT. His research mainly focuses the automated visual inspection of transparent objects. Johannes Meyer works in close cooperation to the Fraunhofer Institute of Optronics, System Technologies and Image Exploitation (IOSB) in Karlsruhe.
Karlsruhe Institute of Technology KIT, Vision and Fusion Laboratory, Karlsruhe, Germany
Thomas Längle is associate professor at the Karlsruhe Institute of Technology (KIT), Karlsruhe and the head of the business unit “Vision Based Inspection Systems” (SPR) at the Fraunhofer IOSB in Karlsruhe, Germany. His research interests include different aspects of image processing and real-time algorithms for inspection systems. He also offers lectures in computer science at the Karlsruhe Institute of Technology and initiates many possibilities for students to work on applied research.
Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB, Karlsruhe, Germany
Jürgen Beyerer is the director of the Fraunhofer Institute of Optronics, System Technologies and Image Exploitation (IOSB) and the head of the Vision and Fusion Laboratory (IES) at the Faculty of Informatics, Karlsruhe Institute of Technology (KIT). His main fields of research are: Automated visual inspection and image processing, fusion of heterogeneous information sources, information theory, system theory, statistical methods and metrology.
Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB, Karlsruhe, Germany
©2016 Walter de Gruyter Berlin/Boston
