Conceptual Framework for Synthetic Data Generation in Remanufacturing
Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › begutachtet
Standard
Production at the Leading Edge of Technology: Proceedings of the 14th Congress of the German Academic Association for Production Technology (WGP), Chemnitz University of Technology, December 2024. Hrsg. / Welf-Guntram Drossel; Steffen Ihlenfeldt; Martin Dix. Springer Nature, 2025. S. 317-325 (Lecture Notes in Production Engineering; Band Part F679).
Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › begutachtet
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - Conceptual Framework for Synthetic Data Generation in Remanufacturing
AU - Koch, Dominik
AU - Stamer, Florian
AU - Lanza, Gisela
N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025
Y1 - 2025
N2 - Remanufacturing requires effective defect detection, but data scarcity and defect variability challenge ML model training. This paper presents a synthetic data generation pipeline focusing on bevel gears, simulating geometric defects like broken teeth, and applying scene randomization to diversify the dataset. The pipeline successfully generates synthetic data for geometric defects, supporting automated segmentation and annotation. However, the current results are limited to geometric defects. Future work will expand to surface and material defects using MDL graphs to improve the pipeline’s applicability. In conclusion, while the pipeline shows potential, real-world validation is necessary to confirm its robustness in remanufacturing.
AB - Remanufacturing requires effective defect detection, but data scarcity and defect variability challenge ML model training. This paper presents a synthetic data generation pipeline focusing on bevel gears, simulating geometric defects like broken teeth, and applying scene randomization to diversify the dataset. The pipeline successfully generates synthetic data for geometric defects, supporting automated segmentation and annotation. However, the current results are limited to geometric defects. Future work will expand to surface and material defects using MDL graphs to improve the pipeline’s applicability. In conclusion, while the pipeline shows potential, real-world validation is necessary to confirm its robustness in remanufacturing.
KW - optical inspection
KW - remanufacturing
KW - synthetical data
KW - Engineering
UR - http://www.scopus.com/inward/record.url?scp=105011695302&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-86893-1_35
DO - 10.1007/978-3-031-86893-1_35
M3 - Article in conference proceedings
AN - SCOPUS:105011695302
SN - 978-3-031-86892-4
SN - 978-3-031-86895-5
T3 - Lecture Notes in Production Engineering
SP - 317
EP - 325
BT - Production at the Leading Edge of Technology
A2 - Drossel, Welf-Guntram
A2 - Ihlenfeldt, Steffen
A2 - Dix, Martin
PB - Springer Nature
ER -