magistrsko delo
Jernej Nejc Dougan (Author), Mojca Kosmatin Fras (Mentor), Dejan Grigillo (Co-mentor), Rok Valič (Co-mentor)

Abstract

Magistrsko delo »Uporaba brezpilotnega sistema za zajem prostorskih podatkov in ocena kakovosti klasifikacije fotogrametričnega oblaka točk« zajema pregled področja brezpilotnih letalnikov za zajem prostorskih podatkov, teorijo obdelave in možnosti klasifikacije fotogrametričnih oblakov točk z oceno kakovosti. V delu smo opredelili brezpilotne sisteme in podali značilnosti brezpilotnih sistemov, namenjenih za zajem prostorskih podatkov. Opredelili smo delitev brezpilotnih sistemov in opisali podsisteme, ki sestavljajo brezpilotne sisteme za uporabo v geodeziji. Na kratko opišemo pravno podlago in zračni prostor v Republiki Sloveniji. Opišemo brezpilotni sistem senseFly eBee, katerega smo tudi uporabili za izvedbo praktičnega dela. Podrobneje opišemo programski paket za obdelavo posnetkov PostFlight Terra 3D. Opišemo delovni postopek za zajem podatkov z brezpilotnim sistemom. V teoretičnem delu podrobno opišemo metode in postopke, ki jih implementirajo programi za obdelavo fotografij, pridobljenih z brezpilotnimi sistemi. Predstavimo, kako je narejena rekonstrukcija in orientacija posnetkov s postopkom Structure from Motion, izravnava bloka posnetkov in samokalibracija. Opišemo eno izmed metod zgoščevanja oblaka točk in pregledamo metode za klasifikacijo oblakov točk. Podamo opise izdelave drugih fotogrametričnih izdelkov kot so DMR in ortomozaik. V okviru praktične raziskave smo izvedli projekt zajema prostorskih podatkov z brezpilotnim sistemom. Načrtovali smo zajem, izvedli zajem podatkov, obdelali pridobljene aeroposnetke, iz njih izdelali oblak točk, DMP in ortomozaik. Brezpilotni sistemi lahko v nekaterih pogledih nadomestijo aerolasersko skeniranje, zato smo izvedli primerjavo med fotogrametričnim oblakom točk in oblakom točk, pridobljenim z aerolaserskim skeniranjem. Fotogrametrični oblak točk smo klasificirali z različnimi metodami in programi. Za oceno kakovosti klasifikacije smo izdelali referenčni oblak točk ter lasten program za ocenjevanje kakovosti klasifikacije. Podali smo ocene kakovosti različni metod klasificiranja. Podamo možnosti za izboljšave klasifikacij fotogrametričnega oblaka točk.

Keywords

geodezija;magistrska dela;brezpilotni sistemi;brezpilotni letalnik;fotogrametrični oblak točk;klasifikacija;ocena kakovosti;prostorski podatki;

Data

Language: Slovenian
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UL FGG - Faculty of Civil and Geodetic Engineering
Publisher: [J. N. Dougan]
UDC: 528.7:629.7 (043.3)
COBISS: 7278177 Link will open in a new window
Views: 2630
Downloads: 770
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Spatial data acquisition with unmanned aerial system and evaluation of photogrammetric point cloud classification quality
Secondary abstract: Master thesis “Spatial data acquisition with unmanned aerial system and evaluation of photogrammetric point cloud classification quality” provides an overview of the field of unmanned aerial systems for spatial data acquisition, theory of data processing and possibilities of automatic photogrammetric point cloud classification with quality evaluation. We define unmanned aerial systems for spatial data acquisition, their characteristic and describe necessary subsystems. We briefly describe the legal basis and the airspace of the Republic of Slovenia. We describe unmanned aerial system senseFly eBee, which was also used to carry out practical work and PostFlight Terra 3D, software for data processing. We define typical workflow for spatial data acquisition with unamend aerial systems. The theoretical part is comprised of the detailed description of the methods and procedures used for data processing and automatic point cloud classification. We describe methods for 3D reconstruction and orientation from images with the procedure Structure from Motion, bundle block adjustment, self-calibration. We describe one of the methods for densifying point cloud and review some methods for automatic classification. In the context of practical research we carried out a project with unamend aerial system. The practical work included survey flight planning, carrying out survey flight and processing of the data. Three main products were created: dense point cloud, digital elevation model and orthomosaic. Photogrammetric unmanned aerial systems may in some aspects replace aerial laser scanning, so we carried out a comparison between photogrammetric point cloud and laser scanning point cloud. The main goal of the practical part of the thesis was to test automatic classification of the photogrammetric point cloud. We performed many classification with different software’s, with different input parameters and evaluate quality of different methods according to ground truth data.
Secondary keywords: geodesy;master of science thesis;unmanned aerial system;unmanned aerial vehicle;photogrammetric point cloud;classification;quality assessment;spatial data;
File type: application/pdf
Type (COBISS): Master's thesis/paper
Thesis comment: Univ. v Ljubljani, Fak. za gradbeništvo in geodezijo
Pages: XX, 84 str., [11] str. pril.
ID: 9064221