Language: | Slovenian |
---|---|
Year of publishing: | 2010 |
Source: | Ljubljana |
Typology: | 2.08 - Doctoral Dissertation |
Organization: | UL FGG - Faculty of Civil and Geodetic Engineering |
Publisher: | [D. Grigillo] |
UDC: | 528.72+528.74:528.8:721(043.3) |
COBISS: | 4890977 |
Views: | 1897 |
Downloads: | 651 |
Average score: | 0 (0 votes) |
Metadata: |
Secondary language: | English |
---|---|
Secondary title: | Automatic building detection from high resolution imagery for maintenance of topographic data |
Secondary abstract: | Topographic data are data on the physical state of the surface. They are managed in topographic databases which need to be regularly updated and maintained. Building data are one of the data types in topographic databases. This doctoral thesis presents data sources and existing methods for management of building data. It notes new data sources that can be used as alternatives to the existing ones. High-resolution image sources enable the use of automatic methods for updating topographic databases. This doctoral thesis reviews the possibilities for maintaining topographic data with single-image methods based on automatic methods. The possibilities for automatic capturing and extraction of buildings are described based on a combination of normalized digital surface model (nDSM) and orthophotos, on a stereo IKONOS imagery and on GeoEye-1 image. Alongside, the thesis presents a new method of automatic building extraction based on an original combination of anisotropic diffusion, unsupervised classification, colour segmentation, and region growing. The methods used are based on building mask creation and its final vectorization. The results of automated building extraction were evaluated. A method of using the results for maintaining topographic building data is proposed. The thesis also proposes further research with the aim of designing a future fully automated system for maintaining topographic building data. High-resolution satellite images enable an important new source for maintaining topographic data. This thesis presents a selection of methods for their georeferencing and methods for enhancing images prior to automatic building extraction. We assessed their suitability for maintaining topographic building data, analysed the geometric accuracy they can provide and highlighted the advantages and challenges of their implementation. X |
Secondary keywords: | building extraction;change detection;nDSM;normalized digital surface model;orthophoto;GeoEye-1;IKONOS;georeferencing;image processing;remote sensing;Zgradbe;Disertacije;Topografski podatki; |
URN: | URN:NBN:SI |
File type: | application/pdf |
Type (COBISS): | Dissertation |
Thesis comment: | Univ. v Ljubljani, Fak. za gradbeništvo in geodezijo |
Pages: | XXIV, 156 str. |
Keywords (UDC): | mathematics;natural sciences;naravoslovne vede;matematika;astronomy;astrophysics;space research;geodesy;astronomija;astrofizika;raziskovanje vesolja;geodesy;surveying;photogrammetry;remote sensing;cartography;geodezija;pregled terena;fotogrametrija;daljinsko zaznavanje;kartografija;photogrammetry: aerial;terrestrial;aerofotogrametrija;terestrična fotogrametrija;mathematics;natural sciences;naravoslovne vede;matematika;astronomy;astrophysics;space research;geodesy;astronomija;astrofizika;raziskovanje vesolja;geodesy;surveying;photogrammetry;remote sensing;cartography;geodezija;pregled terena;fotogrametrija;daljinsko zaznavanje;kartografija;photogrammetry: aerial;terrestrial;aerofotogrametrija;terestrična fotogrametrija;mathematics;natural sciences;naravoslovne vede;matematika;astronomy;astrophysics;space research;geodesy;astronomija;astrofizika;raziskovanje vesolja;geodesy;surveying;photogrammetry;remote sensing;cartography;geodezija;pregled terena;fotogrametrija;daljinsko zaznavanje;kartografija;remote sensing;the arts;recreation;entertainment;sport;umetnost;razvedrilo;zabava;šport;architecture;arhitektura;buildings generally;zgradbe splošno; |
ID: | 26085 |