Multi-Classification of Satellite Imagery Using Fully Convolutional Neural Network

The article considers deep learning techniques, namely, the use of a deep neural network or convolutional neural network (CNN), which increases the efficiency of the application of remote sensing data for multi-classification due to feature learning. In this paper, we have established a classification model using deep convolutional neural networks that can reliably identify the corresponded objects. The explanation of the traditional convolutional neural network and the training process of the proposed convolutional neural network model are presented. The evaluation performances of the proposed model are conducted on the UC Merced Land Use dataset. The proposed model performs high classification accuracy in smallest times without high computation performance.