Power transmission and transformation equipment detection and segmentation method for unmanned inspection in electric power scene

The invention relates to the technical field of power transmission and transformation equipment inspection, in particular to a power transmission and transformation equipment detection and segmentation method for unmanned inspection in an electric power scene, which comprises the following steps of: performing target detection on an obtained picture by adopting a pre-trained detection network model to obtain a semantic type, a detection frame and classification confidence of a target object; a result output by the detection network model is adopted to provide a detection frame area and a corresponding semantic tag for the fine-tuned segmentation large model, the segmentation large model receives the detection frame as a prompt, and the semantic target of the type is accurately segmented in the area in the frame; the method comprises the following steps: preprocessing original data, and encoding a manually labeled detection frame and a segmentation mask into acceptable inputs of a detection model and a segment