Intestinal polyp detection method based on deep supervision and step-by-step learning

The invention discloses an intestinal polyp detection method based on deep supervision and step-by-step learning. The method comprises the following steps: carrying out multi-scale feature extraction on an image shot by an enteroscope by utilizing PVTV2; inputting the extracted multi-scale features into a detail enhancement module, and outputting features after detail enhancement; inputting the features after detail enhancement into a guidance fusion module, and enabling the low-level features to be fused with the high-level features under the guidance of the high-level features; inputting the features of the highest layer into a multi-branch decoder module to obtain features after multi-scale information fusion decoding; inputting the features after detail enhancement and the output result of the previous multi-scale decoder into the multi-scale decoder, and gradually learning to obtain the output features of each layer of multi-scale decoder; and the output features of the multi-scale decoder pass through c