Lightweight Convolutional Neural Network for High-Spatial-Resolution Remote Sensing Scenes Classification

High spatial resolution remote sensing (HSRRS) image classification and detection are playing an increasingly important role in land planning and resource utilization. At present, convolutional neural networks (CNN) have made great achievements in computer vision and natural language processing. CNN is also widely used in various tasks such as HSRRS image classification and detection. However, existing neural networks generally have the redundant parameters and the slow computation speeds, which are inconvenient to implement them into the mobile terminals. To solve this problem, in this paper, we propose a lightweight CNN model, which is named as the MobileNet. Simulation results are given to confirm that MobileNet can greatly reduce the parameters and accelerate the computation while ensuring high accuracy.