CPC G06T 7/0012 (2013.01) [G06V 10/24 (2022.01); G06V 10/761 (2022.01); G06V 10/82 (2022.01); G06T 2207/10116 (2013.01)] | 19 Claims |
1. A method for generating synthetic X-ray images, the method comprising:
providing a trained first neural network for generating at least one synthetic X-ray image of specified quality, and a trained second neural network for providing characterizing properties from at least one secondary X-ray image for the first neural network, wherein the first neural network and the second neural network are trainable by primary X-ray images each having a specified minimum quality;
ascertaining the characterizing properties using the at least one secondary X-ray image that has a lower quality compared to the primary X-ray images, by the second neural network; and
generating the at least one synthetic X-ray image by the first neural network using the characterizing properties provided by the second neural network, the at least one synthetic X-ray image being improved with respect to quality compared to the at least one secondary X-ray image.
|