KIT | KIT-Bibliothek | Impressum | Datenschutz

Domain-Shift-Dataset of Defects on Metallic Surfaces (MSD-Shift)

Schlagenhauf, Tobias 1
1 Institut für Produktionstechnik (WBK), Karlsruher Institut für Technologie (KIT)


Zugehörige Institution(en) am KIT Institut für Produktionstechnik (WBK)
Publikationstyp Forschungsdaten
Publikationsdatum 29.06.2022
Erstellungsdatum 01.04.2022 - 06.06.2022
Identifikator DOI: 10.5445/IR/1000147763
KITopen-ID: 1000147763
Lizenz Creative Commons Namensnennung – Nicht kommerziell – Keine Bearbeitungen 4.0 International
Vorab online veröffentlicht am 08.06.2022
Schlagwörter Transfer Learning, Domain Shift, Domain Generalization, Domain Adaption, Machine Learning, Deep Learning, Metallic Surfaces, Defects
Liesmich

The dataset maps two different surfaces (domains) from mechanical engineering (Surfaces of Ball Screw Drives (BSD); Surface of Metallic Semi-finished Products (SEV)). The domains each contain image data with and image data without surface defects. The surfaces differ, but the defect features are similar across the domains. The dataset is thus suitable for investigating questions in the context of domain shift, domain generalization, and transfer learning. The dataset is structured as follows: BSD with defect (5240); BSD without defect (1896) SEV with defect (2018); SEV without defect (21806). Attention: The Domains are not balanced over classes. The SEV images are each 224x224 pixel RGB PNG files. THE BSD images are each 150x150 pixel RGB PNG files. The SEV data are excerpts from the Severstal dataset.

Art der Forschungsdaten Dataset
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page