Author:
Jeziorski Natascha,Redenbach Claudia
Abstract
AbstractTraining defect detection algorithms for visual surface inspection systems requires a large and representative set of training data. Often there is not enough real data available which additionally cannot cover the variety of possible defects. Synthetic data generated by a synthetic visual surface inspection environment can overcome this problem. Therefore, a digital twin of the object is needed, whose micro-scale surface topography is modeled by texture synthesis models. We develop stochastic texture models for sandblasted and milled surfaces based on topography measurements of such surfaces. As the surface patterns differ significantly, we use separate modeling approaches for the two cases. Sandblasted surfaces are modeled by a combination of data-based texture synthesis methods that rely entirely on the measurements. In contrast, the model for milled surfaces is procedural and includes all process-related parameters known from the machine settings.
Funder
Bundesministerium für Bildung und Forschung
Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM
Publisher
Springer Science and Business Media LLC
Reference39 articles.
1. Bechikh A, Klinkova O, Maalej Y, Tawfiq I, Nasri R. Sandblasting parameter variation effect on galvanized steel surface chemical composition, roughness and free energy. Int J Adhes Adhes. 2020;102:102653.
2. Bosnar L, Gospodnetić P. Procedural defect modeling for virtual surface inspection environments. IEEE Comput Graph Appl. 2023:43(2):13–22.
3. Bosnar L, Rauhut M, Hagen H, Gospodnetić P. Texture synthesis for surface inspection. In: LEVIA 22: Leipzig symposium on visualization in applications. 2022.
4. Bosnar L, Šarić D, Dutta S, Weibel T, Rauhut M, Gospodnetić P. Image synthesis pipeline for surface inspection. In: LEVIA’20: Leipzig symposium on visualization in applications 2022. 2020.
5. de Lacalle LNL, Campa FJ, Lamikiz A. 3 - milling. In: Paulo Davim J, editor. Modern machining technology. Woodhead Publishing; 2011. p. 213–303.