A Lightweight Convolutional Neural Networks for Scene Classification

With the rapid advancements in deep learning and remote sensing, the classification of remote sensing scenes faces significant challenges due to the high consumption of computational resources. To tackle this issue, this paper proposes the utilization of the MobileNetV2 neural network based on transfer learning for remote sensing scene classification, which enhances efficiency while maintaining classification accuracy. Experimental results show that the model achieves a classification accuracy of 94.57% on the UC_Merced dataset, significantly outperforming AlexNet, ResNet18, VGG11, and DenseNet121.