Unet network brain tumor MRI image segmentation method for improving attention module

The invention discloses a Unet network brain tumor MRI image segmentation method for improving an attention module, the problem that due to the diversity of shapes of lesions and the difference of different organ structures, the requirements for accuracy, speed and the like cannot be met by only using a UNet structure to segment the lesions is solved, and the invention belongs to the field of semantic segmentation. The method comprises the following steps: replacing an original convolution module with a reversible residual module in a Unet network; adding an improved residual attention module ResCBAM to a jump connection part in the Unet before an encoder and a deencoder are spliced, so that the quality of detail features in a training process can be better improved. An experiment is carried out by using a data set of Brats2019, and an experiment result shows that compared with other methods, the method provided by the invention has the advantage that the segmentation effect is improved to different degrees u