abstract |
The present application provides an image classification model training method, image classification method and device, including: obtaining remote sensing images of a preset area before and after the change of land types; Band enhancement to obtain the first target image and the second target image; based on the first target image and the second target image, respectively, construct a two-class training set and a multi-class training set; based on the two-class training set and the multi-class training set, the transfer The subsequent pre-training model is trained until the loss function of the pre-training model reaches the preset convergence condition, and the image two-classification model and the image multi-classification model are obtained. The image two-classification model is used for image change detection, and the image multi-classification model is used. Class classification. The invention can enrich the band features of the remote sensing images, improve the model accuracy and reduce the difficulty of sample collection. |