| abstract |
The invention discloses an optical remote sensing image target detection system based on a deep convolutional neural network, comprising an image acquisition module for acquiring optical remote sensing images of ships; and a preprocessing module for degrading the optical remote sensing images of ships Noise, land and sea segmentation and edge extraction; sample division module, used to divide preprocessed optical remote sensing images into training images and validation images; attention module, used to extract better features from preprocessed optical remote sensing images Representation; the target detection module is used to construct and train a deep convolutional neural network model, and after obtaining the trained deep convolutional neural network model, target detection is performed on optical remote sensing images. The invention increases the attention module, improves the efficiency of the optimization model, and effectively improves the detection accuracy of the optical remote sensing image ship. |