TTK Links: Home Page | TTK Moodle | ÕIS
TTK/UAS Repository For Digital Teaching and Learning Resources

Maapealse laserskaneerimise mudeldamise täpsuse määramine

Kuurberg, Kaur (2014) Maapealse laserskaneerimise mudeldamise täpsuse määramine. [thesis] [en] Modelling accuracy of terrestrial laser scanning.

[img]
Preview
PDF - Published Version
Download (1MB) | Preview
[img]
Preview
Image (JPEG) (lihtlitsents) - Supplemental Material
Download (1MB) | Preview

Abstract

Antud lõputöö eesmärgiks oli uurida maapealse laserskaneerimise andmete mudeldamise täpsust. Selle määramiseks kasutati puidust kasti ja samott-toru. Skaneeritavate objektide valiku kriteeriumiks oli, et üks oleks ümar ning teine kandiliste servadega. Skaneerimisprotsess viidi läbi Tallinna Tehnikakõrgkooli keldrikorrusel asuvas ligi 80 m pikkuses koridoris. Kasti ja toru skaneeriti neljalt erinevalt kauguselt erinevate resolutsioonidega. Kokku teostati ühest jaamast 3 skaneerimist. Igal kaugusel asetsesid skaneeritavad objektid laserskanneri suhtes alati sama nurga all. Selleks märgiti nad enne elektrontahhümeetriga täpselt maha. Saadud punktipilved mudeldati kolme erineva tarkvaraga. Kasutati Leica Geosystems AG Cyclone 8.0, Autodesk Inc Revit 2015 ning Technodigit 3D Reshaper 2013 tarkvarasid. Toodi välja saadud mudelite täpsuse erinevused ning selgitati tarkvarade tööpõhimõtteid. Samuti anti töös ülevaade, kui kvaliteetset punktipilve mingi resolutsioon teatud kauguselt võimaldab. Selgus ka millistelt kauguselt alates ei ole madalamaid resolutsioone enam mõtet kasutada, kuna midagi ei jää peale. Töö lõpuosas selgitati välja, milline on keskmine mudeldamise täpsus kindla resolutsiooniga kindlalt kauguselt, et oleks võimalik hinnata, millal on vaadeldud resolutsioone kõige otstarbekam kasutada. Igalt kauguselt kõige kõrgema resolutsiooniga saadud mudelist tehti ka lõiked ning viidi läbi student’i t-test. Student’i test näitab, kas ühe lõike hälbed erinevad oluliselt teiste sama mudeli lõigete hälvetest ning mõjutavad seeläbi mudeli täpsust. Uurimisel seatud eesmärgid täideti. Selgitati välja, kui palju sõltub punktipilvest loodud mudeli täpsus kasutatud tarkvara valikust. Leiti ka, milliseid resolutsioone ning millistel tingimustel oleks kõige otstarbekam kasutada.

Abstract [en]

Modelling accuracy of terrestrial laser scanning. The main goal of this research is to investigate how modelling accuracy changes when the same objects are scanned at different distances using different resolutions. A wooden box and a fire clay pipe were used. True sizes of the box and the pipe were measured by calliper. The scanning process took place in the basement of Tallinn University of Applied Sciences. There was a corridor about 2 m wide and 80 m long. The objects were scanned at 10 m, 30 m, 60 m and 77,35 m distances. Leica Scanstation C10 laserscanner was used. Scanner position did not change during the process. Only the scanned objects were moved. The objects at distances 10 m and 30 m were scanned using low, medium and high resolutions. At the distance of 60 m all four available resolutions were used and at 77,35 m medium, high and the highest resolutions were used. All point clouds were modelled with three different softwares. Leica Cyclone 8.0, Autodesk Inc Revit 2015 and 3D Reshaper from Technodigit were used. After modelling the object sizes were determined based on the models. The deviations from true values were also calculated. The results show the quality of point clouds from every distance and indicate the best resolution to be used at different distances. It was also found that the scanning the box needs a higher resolution than pipe to get the models with same accuracy. A Student’s t-test with the scanning results was also carried out. Only the results of the 3D Reshaper were used in the test. For that cross-sections were made from both objects. The test was conducted to show whether the measured lengths from the model and the true object were consistent with each ohter or not.

Item Type: thesis
Advisor: Tarvo Mill
Subjects: Construction > Applied Geodesy > Geodesy > Theory of measurement error
Construction > Applied Geodesy > Geodesy > Geodetic instruments
Divisions: Institute of Construction > Applied Geodesy
Depositing User: Kaur Kuurberg
Date Deposited: 19 Jun 2014 12:07
Last Modified: 19 Jun 2014 12:08
URI: http://eprints.tktk.ee/id/eprint/738

Actions (login required)

View Item View Item