Small sample segmentation method and system combining hierarchical sparse representation and adaptive prompt

The invention belongs to the technical field of computer vision, and particularly discloses a small sample segmentation method and system combining hierarchical sparse representation and adaptive prompt, and the method comprises the following steps: obtaining an image set of a support image and a query image; using a fixed backbone to extract multi-scale image features of the support image and the query image; inputting the extracted multi-scale image features into a hierarchical sparse representation module (HSRM) and an adaptive prompt meta-learner (APM) at the same time, applying sparse constraint to the multi-scale image features, and generating a category graph; the generated category graph is input into a decoder module, predicting a final segmentation of the query image. By adopting the technical scheme, a more accurate pixel relationship between the support image and the query image is mined, the transmission of category information is enhanced, and the generalization ability of the model is improved.