Convolutional neural networks for global human settlements mapping from Sentinel-2 satellite imagery

Abstract

Spatially consistent and up-to-date maps of human settlements are crucial for addressing policies related to urbanization and sustainability, especially in the era of an increasingly urbanized world. The availability of open and free Sentinel-2 data of the Copernicus Earth Observation program offers a new opportunity for wall-to-wall mapping of human settlements at a global scale. This paper presents a deep-learning-based framework for a fully automated extraction of built-up areas at a spatial resolution of 10 m from a global composite of Sentinel-2 imagery. A multi-neuro modeling methodology building on a simple Convolution Neural Networks architecture for pixel-wise image classification of built-up areas is developed. The core features of the proposed model are the image patch of size 5 × 5 pixels adequate for describing built-up areas from Sentinel-2 imagery and the lightweight topology with a total number of 1,448,578 trainable parameters and 4 2D convolutional layers and 2 flattened layers. The deployment of the model on the global Sentinel-2 image composite provides the most detailed and complete map reporting about built-up areas for reference year 2018. The validation of the results with an independent reference dataset of building footprints covering 277 sites across the world establishes the reliability of the built-up layer produced by the proposed framework and the model robustness. The results of this study contribute to cutting-edge research in the field of automated built-up areas mapping from remote sensing data and establish a new reference layer for the analysis of the spatial distribution of human settlements across the rural–urban continuum.

Gridded population mapping for Germany based on building density, height and type from Earth Observation data using census disaggregation and bottom-up estimates

“…Current and future developments in directly mapping high-resolution building density from Earth Observation data, e.g. the convolutional neural network based approach in [ 65 ], could contribute to a more reliable distinction of building and non-building impervious surface density without additional OSM data. A building type layer was specifically created for this study.…”

Analysis of Built-Up Areas of Small Polish Cities with the Use of Deep Learning and Geographically Weighted Regression

“…A certain problem in determining the built-up areas in all cities in a given year is the lack of available up-to-date geodetic data sources. This is made possible by information extracted from satellite images [35,[89][90][91]. The aim of this study was to investigate the spatial differentiation of the share of built-up areas in the area of small cities in Poland, as well as to search for factors influencing this differentiation.…”

Generalized Vertical Components of built-up areas from global Digital Elevation Models by multi-scale linear regression modelling

“…For instance Sentinel satellite sensors could be used to generate data composites characterizing vegetation presence from calibrated reflectance data [97] or characterizing the presence of built-up areas [98][99][100]. The delineation of built-up areas from both Sentinel 1 radar data [99] and Sentinel 2 optical data [98,100] has a higher spatial resolution as compared to the Landsat data tested in this study. Consequently, an improvement would be expected especially regarding the estimation of net vertical components of built-up areas, which are more dependent on the spatial resolution than the gross vertical components.…”