Satellite remote sensing image multi-target detection method and device based on improved Cascade R-CNN model, computer program product and computer readable storage medium

The invention discloses a satellite remote sensing image multi-target detection method and device based on an improved Cascade R-CNN model, a computer program product and a computer readable storage medium. The method comprises the following steps of: 1, carrying out image preprocessing on a high-resolution satellite remote sensing image; 2, carrying out image global feature extraction on the preprocessed high-resolution satellite remote sensing image based on an improved feature extraction network; a third step of applying a sliding window mode on the generated global feature map, and generating a target candidate box through a region candidate network; 4, performing interpolation operation on the global feature map in the target candidate box to generate a candidate region global feature map with a fixed size; a fifth step of performing feature extraction on the global feature map of the candidate region based on the improved feature extraction network; and a sixth step of performing classification and regr