Methods of obtaining geospatial data using satellite communications and their processing using convolutional neural networks

P. 1–5.

Tsvetkovskaya I. I., Tekutieva N. V., Prokofyeva E. N., Vostrikov A. V.

The availability of high-resolution satellite images obtained through space radio communications offers the opportunity to use the most advanced technologies and techniques for analyzing remote sensing data. The paper discusses the data obtained with the use of ground-based, airborne or space-based filming equipment, which makes it possible to obtain images in one or several sections of the electromagnetic spectrum. This article provides an overview of existing artificial spacecraft and systems for obtaining space data. Also, there are the examples of the use of convolutional neural networks (CNN) for processing data obtained from artificial Earth satellites. CNN has a high learning ability and the capacity to automatically learn optimal functions based on the data.

Keywords: распознавание образов, классификация, pattern recognition, Спутник, classification, convolutional neural networks, satellite, сверточные нейронные сети, Earth remote sensing, дистанционное зондирование Земли, satellite images, космоснимки

Publication based on the results of: Разработка методики непрерывного мониторинга для выявления экзогенных процессов на инфраструктурных объектах (2021)

In book

2020 Moscow Workshop on Electronic and Networking Technologies (MWENT)

IEEE, 2020.