Remote sensing image scene classification based on residual dense network

To address the problem that remote sensing scenes are semantically rich, texturally complex, and similar scenes are prone to misclassification. In this paper, we propose a method for remote sensing image scene classification based on residual dense network. The method combines the residual dense module, residual module to extract the feature details of remote sensing images, which can effectively improve the scene classification accuracy of remote sensing images. Experiments on the publicly available SIRI-WHU remote sensing scene dataset validated the performance of the algorithm, and the classification accuracy of the method in this paper reached 96.66%. The results show that the classification accuracy of this method is better than GoogeLeNet, ResNet-34 and DenseNet-121 methods.