Home // PATTERNS 2023, The Fifteenth International Conference on Pervasive Patterns and Applications // View article

Spatiotemporal Modeling of Urban Sprawl Using Machine Learning and Satellite Data

Authors:
Alexander Troussov
Dmitri Botvich
Sergey Maruev

Keywords: Earth remote sensing; urbanization; evolutionary models of urban development; cellular automata; machine learning.

Abstract:
The paper discusses the issues related to the use of machine learning in designing and creating predictive models of the territorial development of cities. Special attention is paid to models that use satellite data, which is the most easily accessible type of data and allows for the use of machine learning. The proposed provisions are based on the authors’ current experience in developing an information system for forecasting the territorial development of cities based on remote sensing data of the Earth. A novel approach to developing predictive models of urban growth is presented, which is based on machine learning methods. The approach uses satellite data from the Defense Meteorological Satellite Program/Operational Linescan System and Visible Infrared Imaging Radiometer Suite/Day-Night Band for training. The corresponding machine learning optimization problem is presented and discussed.

Pages: 52 to 56

Copyright: Copyright (c) IARIA, 2023

Publication date: June 26, 2023

Published in: conference

ISSN: 2308-3557

ISBN: 978-1-68558-049-0

Location: Nice, France

Dates: from June 26, 2023 to June 30, 2023