Intestinal polyp segmentation method and system fused with mixed attention mechanism, and medium

The invention discloses an intestinal polyp segmentation method and system fused with a mixed attention mechanism, and a medium, and the method comprises the following steps: carrying out the preprocessing of an intestinal polyp image based on an endoscope, including data enhancement, and region extraction of a segmentation target through an adaptive threshold value, and is used for enhancing the robustness of a segmentation model and mining more accurate boundary information of the target; a feature token pyramid module is constructed to improve the semantic information extraction capability of an intestinal polyp image, and meanwhile, semantic information with multi-scale perception can be obtained by constructing a feature token pyramid by adopting fewer modules; a global feature extraction module is constructed, a local-global training strategy is utilized to reduce the requirement of the segmentation model for the data sample size, and the segmentation performance is further improved; a feature injection