Optical microscopy images of cast iron alloys defects

The dataset comprises two subdatasets, featuring images of carbon defects based on cast iron samples adhering to European standards. The images are in .png format and are divided into training and test sets, with an additional split into 5 folds for reproducibility.The first sub-dataset contains ima...

Descripción completa

Detalles Bibliográficos
Autores: Bárcena Rodríguez, Marta, Lloret Iglesias, Lara, Ferreño, Diego, Carrascal Vaquero, Isidro
Tipo de recurso: conjunto de datos
Fecha de publicación:2022
País:España
Institución:Consejo Superior de Investigaciones Científicas (CSIC)
Repositorio:DIGITAL.CSIC. Repositorio Institucional del CSIC
OAI Identifier:oai:digital.csic.es:10261/363741
Acceso en línea:http://hdl.handle.net/10261/363741
https://doi.org/10.20350/digitalCSIC/16484
Access Level:acceso abierto
Palabra clave:Deep learning
Image classification
Iron casting
Microscopic imaging
Descripción
Sumario:The dataset comprises two subdatasets, featuring images of carbon defects based on cast iron samples adhering to European standards. The images are in .png format and are divided into training and test sets, with an additional split into 5 folds for reproducibility.The first sub-dataset contains images with only one type of defect per image. The second sub-dataset combines defects from consecutive categories to more closely represent real-world scenarios. Additionally, the second sub-dataset includes the corresponding semantic segmentation masks (MASK) as well as pre-trained models (.h5) for classification purposes.