Immunofluorescent staining neuron cell body segmentation method based on semi-supervised and diffusion model

The invention relates to the technical field of data processing, in particular to an immunofluorescence staining neuron cell body segmentation method based on a semi-supervised and diffusion model, comprising the following steps: S10, training a diffusion model and a semi-supervised model based on the same original data set; s20, using the semi-supervised model obtained by training in the step S10 to segment the unmarked cell body image, generating a pseudo label, inputting the image containing the pseudo label into the diffusion model obtained by training in the step S10 to generate a corresponding pseudo image, and forming a pseudo data set by the pseudo image; and S30, retraining the semi-supervised model trained in S10 based on the pseudo data set and the original data set, and segmenting the immunofluorescence staining neuron cell body by using the trained semi-supervised model. According to the method, the sample data is supplemented by generating the pseudo image, so that the dependence on the sample d