Key points are not available for this paper at this time.
Remote sensing technology is a powerful tool for a wide range of applications, from medical diagnoses to environmental monitoring. Quality inspection, inventory management, environmental monitoring, supply chain analysis, and predictive maintenance are just a few of the many industrial uses for remote sensing image classification using deep learning. It's a tool for lowering production costs without sacrificing quality or long-term viability. Remote sensing image classification with deep learning aids production and sustainability by offering datadriven decision-making and useful insights. In this paper, we review the application of deep learning techniques in the field of remote sensing data analysis. This paper aims to investigate several techniques for visualising model decisions and to attribute them to specific aspects within the dataset. The proposed techniques include deep convolutional neural networks (CNNs) with saliency stream and RGB stream fusion techniques. In addition, we also discuss the use of extreme learning machine (ELM) classifiers with fused features as input for results. Finally, we discuss the performance of the proposed techniques on the UC Merced Land-Use dataset, Aerial Image dataset (AID), and NWPU-RESISC45 datasets. The results of the experiments demonstrate that the proposed techniques outperform other existing techniques. Additionally, the fused features from different streams improve the performance of the model significantly. This paper focused on information on various related research works and their models, including datasets. The main purpose is to make the already existing bridge between social life and the computer system even more robust.
Tarun et al. (Thu,) studied this question.