magistrsko delo
Veljko Miljković (Author), Tilen Urbančič (Mentor)

Abstract

V magistrskem delu obravnavamo trenutno dostopna orodja samodejnega prepoznavanja linij v oblakih točk, pridobljenih s terestričnim laserskim skeniranjem. Hiter razvoj tehnologij za zajem podatkov in njihova naraščajoča uporaba v geodeziji, gradbeništvu in drugih panogah ustvarjata potrebo po še vedno točni, a hitrejši obdelavi podatkov. V nalogi smo preučili in primerjali nekatere izmed komercialnih orodij, trenutno dostopnih na trgu: Civil 3D, ReCap, UNDET in CloudCompare. Za izvedbo testiranja in analize smo pripravili dva referenčna izrisa, Most in Stavba, na podlagi katerih smo izvajali primerjave. Testna primera sta reprezentativna vzorca notranjih prostorov in zunanjosti inženirskega objekta, da bi tako lahko preučili vpliv različnih pogojev na delovanje algoritmov. Rezultati so pokazali, da se orodja razlikujejo glede na zmogljivosti pri prepoznavanju linij v dveh in treh dimenzijah ter glede na njihovo natančnost, zanesljivost, hitrost in uporabnost. Nobeno izmed orodij se ni izkazalo kot univerzalno uporabno za oba testna primera, kar kaže na omejitve posameznih pristopov. Poleg tehničnih vidikov je raziskava vključila tudi stroškovno analizo orodij. Na podlagi ugotovitev so bila orodja ocenjena, pri čemer so izpostavljene njihove prednosti in pomanjkljivosti glede na specifične potrebe in pogoje uporabe.

Keywords

magistrska dela;geodezija;oblak točk;linije;samodejna prepoznava;terestrično lasersko skeniranje;

Data

Language: Slovenian
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UL FGG - Faculty of Civil and Geodetic Engineering
Publisher: [V. Miljković]
UDC: 528.7:004(043.2)
COBISS: 227426307 Link will open in a new window
Views: 32
Downloads: 4
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: Automated 3D line detection in a point cloud
Secondary abstract: In this master thesis, we discuss the currently available tools for automatic line detection in point clouds obtained from terrestrial laser scanning. The rapid development of data acquisition technologies and their increasing use in surveying, construction and other industries creates a need for still accurate but faster data processing. In this thesis we have examined and compared some of the commercial tools currently available on the market: Civil 3D, ReCap, UNDET and CloudCompare. In order to carry out the testing and analysis, two reference drawings, Bridge and Building, were prepared and used as a basis for the comparisons. The test cases are representative samples of the interior and exterior of an engineering building, in order to study the impact of different conditions on the performance of the algorithms. The results showed that the tools differed in terms of their performance in line detection in two and three dimensions, as well as in terms of their accuracy, reliability, speed and usability. None of the tools proved to be universally applicable to both test cases, indicating the limitations of each approach. In addition to the technical aspects, the study also included a cost analysis of the tools. Based on the findings, the tools were evaluated, highlighting their strengths and weaknesses according to the specific needs and conditions of use.
Secondary keywords: master thesis;geodesy;point cloud;lines;line extraction;terrestrial laser scanning;
Type (COBISS): Master's thesis/paper
Thesis comment: Univ. v Ljubljani, Fak. za gradbeništvo in geodezijo
Pages: 1 spletni vir (1 datoteka PDF (VIII, 41 str.))
ID: 25974258