2D images of CT scans of carbon fiber reinforced polyamide 6

They were generated by augmenting (x4) and cutting (x4) of 1830 single 2D images resulting from 18 different specimens of 2 CFRP plaques. The images were used as training data for a generative adversarial network (GAN) which creates similar microstructural images.

Please find details in the following publication: Blarr, J.; Klinder, S.; Liebig, W. V.; Inal, K.; Kärger, L.; Weidenmann, K. A.: Deep convolutional generative adversarial network for generation of computed tomography images of heterogeneous low-contrast microstructures. Nature Scientific Reports XXX (2024).