Accurate spatial population data are an important requirement in many applications. In this thesis, the problem of disaggregating the spatial distribution of population density and rent costs using a machine learning model is studied. An approach based on freely available ancillary data such as OpenStreetMap and Urban Atlas is proposed and implemented in the form of an automated Python toolbox for ArcGIS. The applications on the urban areas of Prague, Vienna and Ljubljana show promising results, overperforming the competing population disaggregation solutions in spatial resolution and displaying a satisfying degree of transferability. A number of further improvements is suggested. Powered by TCPDF (www.tcpdf.org)

The topic of this study is to evaluate data sources from the perspective of the information about landscape they provide to their users. The aim is comparison of thematic content of Urban Atlas database and data from Czech cadastre of real estate in Prague metropolitan region between years 2006 and 2012 with focus at meaning of classification systems used by both datasets. The data are processed by evaluation of thematic similarity and statistical tools which quantify similarity between researched data. Results are further verified by using validation data. Important results are visualized by charts, tables and maps. The areas with high degree of dissimilarity were found using chosen methods and their thematic characteristics were further examined as well as their major causes. It was proved that differences between both datasets are significant and they share certain characteristics. It was also proved that cadastral data are to high extent out-of-date. Keywords Urban Atlas, cadastre of real estate, land use, land cover, Prague

Remote Sensing (RS) technology provides valuable information about Earth by enabling an overview of the planet from above, making it a much-needed resource for many applications. Given the abundance of RS data and continued urbanisation, there is a need for efficient approaches to leverage RS data and its unique characteristics for the assessment and management of urban areas. Consequently, employing Deep Learning (DL) for RS applications has attracted much attention over the past few years. In this thesis, novel datasets consisting of satellite RS images over urban areas in Sweden were compiled from Sentinel-2 multispectral, Sentinel-1 Synthetic Aperture Radar (SAR) and Urban Atlas 2018 Land Use/Land Cover (LULC) data. Then, DL was applied for multiband and multiclass semantic image segmentation of LULC. The contributions of complementary spectral, temporal and SAR data and spectral indices to LULC classification performance compared to using only Sentinel-2 data with red, green and blue spectral bands were investigated by implementing DL models based on the fully convolutional network-based architecture, U-Net, and performing data fusion. Promising results were achieved with 25 possible LULC classes. Furthermore, almost all DL models at an overall model level and all DL models at an individual class level for most LULC classes benefited from complementary satellite RS data with varying degrees of classification improvement. Additionally, practical knowledge and insights were gained from evaluating the results and are presented regarding satellite RS data characteristics and semantic segmentation of LULC in urban areas. The obtained results are helpful for practitioners and researchers applying or intending to apply DL for semantic segmentation of LULC in general and specifically in Swedish urban environments.