Deep learning segmentation method of carbon fiber composite material data set

The invention provides a deep learning segmentation method for a carbon fiber composite material data set, and the method comprises the steps: obtaining an image of a related carbon fiber composite material through an XCT technology, constructing an original data set of a model, generating a two-dimensional random synthesis carbon fiber image through parametric modeling, and carrying out the segmentation of the two-dimensional random synthesis carbon fiber image. And generating a virtual data set which is the same as a real structure through an adversarial learning style migration network Pix2PixHD, taking the original data set and the virtual data set as a mixed data set, inputting the mixed data set into a semantic segmentation network based on Swin-Transformer, and finally forming a semantic segmentation network capable of accurately segmenting a carbon fiber composite material image. The method solves the problems that original data sets are insufficient and manual annotation is difficult, intelligent and