Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Kvalitetsbeskrivning och kvalitetspåverkande faktorer vid produktion av TIN-modeller

Andersson, Axel LU (2024) In Examensarbete i geografisk informationsteknik EXTM05 20241
Surveying (M.Sc.Eng.)
Dept of Physical Geography and Ecosystem Science
Abstract (Swedish)
Under de senaste åren har det skett stor utveckling inom teknologin för att samla in höjddata
över stora markområden med bland annat laserskanner. En ökad användning av höjddata i
samhället kombinerat med att arbetsprocesser utvecklats av olika aktörer innebär ett behov av
enhetlig datahantering. I diskussioner om kvalitet på höjddata talas ofta om kvalitet med
avseende på insamlade data och på slutlig produkt. Mindre fokus har däremot lagts på vad som
händer under själva produktionsprocessen.
Detta arbete syftar till att utreda hur kvalitet i triangulerade höjdmodeller, TIN-modeller, ska
beskrivas. Dessutom utreds vilka kvalitetspåverkande faktorer som finns i
produktionsprocessen och deras effekter. Slutligen utreds hur... (More)
Under de senaste åren har det skett stor utveckling inom teknologin för att samla in höjddata
över stora markområden med bland annat laserskanner. En ökad användning av höjddata i
samhället kombinerat med att arbetsprocesser utvecklats av olika aktörer innebär ett behov av
enhetlig datahantering. I diskussioner om kvalitet på höjddata talas ofta om kvalitet med
avseende på insamlade data och på slutlig produkt. Mindre fokus har däremot lagts på vad som
händer under själva produktionsprocessen.
Detta arbete syftar till att utreda hur kvalitet i triangulerade höjdmodeller, TIN-modeller, ska
beskrivas. Dessutom utreds vilka kvalitetspåverkande faktorer som finns i
produktionsprocessen och deras effekter. Slutligen utreds hur specifikationer om datamängder
som punktmoln och brytgeometrier påverkar kvaliteten i TIN-modeller.
Rapporten inleds med en litteraturöversikt för att etablera centrala begrepp och koncept.
Arbetets frågeställningar utreds därefter genom en intervjustudie och fallstudier. I intervjufasen
intervjuas personer som använder TIN om deras användning och krav på TIN-modeller.
Intervjuerna pekar på att datareduktion är en viktig faktor för att effektivisera produktionen av
TIN-modellen och sedan ha en modell som är liten nog att hantera.
I fallstudierna studeras effekten av datareduktion, som framkom i intervjuerna som en viktig
faktor i produktionsprocessen. Dessutom studeras effekten av användning av brytgeometrier.
Fallstudierna visar att beroende på olika faktorer kan stora delar av den ursprungliga
datamängden reduceras bort. I vissa fall räcker det med 1 % av den ursprungliga datamängden
för att skapa acceptabla TIN-modeller. Användande av brytgeometrier kan även möjliggöra
ytterligare datareduktion om de är välplacerade och väl inmätta.
I slutet av rapporten diskuteras resultaten och slutsatser dras. Närmare bestämt diskuteras hur
kvalitet av TIN-modeller kan beskrivas. Dessutom diskuteras det hur datereduktion och andra
parametrar i indata och produktionsprocessen av TIN-modeller påverkar den slutliga kvaliteten. (Less)
Abstract
During recent years there has been great development in the technology used to gather elevation
data over large areas with laser scanners. Increased usage of elevation data in society as well as
work processes being developed simultaneously at different places implies a need for uniform
data management. In discussions regarding elevation data quality, the main issues are often
quality in source data or the finished elevation model. Less attention is however given to what
happens in the production process of elevation models.
This thesis project aims to investigate how quality in triangulated elevation models, TINmodels, should be described. Furthermore, what quality affecting factors exist in the production
process and their effects... (More)
During recent years there has been great development in the technology used to gather elevation
data over large areas with laser scanners. Increased usage of elevation data in society as well as
work processes being developed simultaneously at different places implies a need for uniform
data management. In discussions regarding elevation data quality, the main issues are often
quality in source data or the finished elevation model. Less attention is however given to what
happens in the production process of elevation models.
This thesis project aims to investigate how quality in triangulated elevation models, TINmodels, should be described. Furthermore, what quality affecting factors exist in the production
process and their effects is researched. Lastly, this thesis studies if specifications for data sets
like point clouds and break geometries affect the quality of TIN-models.
The report begins with a literature review to establish central terms and concepts. The research
questions are then studied by conducting an interview study as well as case studies. In the
interview study people that use TIN-models are interviewed about their usage and quality
demands of TIN-models. The interviews indicate that data reduction is an important factor to
optimize production of TIN-models and afterwards have TIN-models small enough to manage.
In the case studies the effects of data reduction, which appeared as an important factor in the
production process, are studied. In addition, the effects of using break geometries are evaluated.
The case studies show that, depending on different factors, large parts of the original data sets
can be reduced. In some cases, less than 1 % of original dataset is needed for creating acceptable
TIN-models. Use of break geometries can also allow for further data reduction if they are well
placed and well produced.
Towards the end of the report the results are discussed and conclusions are drawn. It is discussed
how quality of TIN-models can be described. Additionally, it is discussed how data reduction
and other parameters in source data and production process affect the quality of the final TINmodel. (Less)
Please use this url to cite or link to this publication:
author
Andersson, Axel LU
supervisor
organization
course
EXTM05 20241
year
type
H2 - Master's Degree (Two Years)
subject
keywords
TIN, Markhöjdmodell, Datarduktion, Brytgeometrier, Geodatakvalitet
publication/series
Examensarbete i geografisk informationsteknik
report number
38
language
Swedish
id
9172323
date added to LUP
2024-08-26 12:30:38
date last changed
2024-08-26 12:30:38
@misc{9172323,
  abstract     = {{During recent years there has been great development in the technology used to gather elevation
data over large areas with laser scanners. Increased usage of elevation data in society as well as
work processes being developed simultaneously at different places implies a need for uniform
data management. In discussions regarding elevation data quality, the main issues are often
quality in source data or the finished elevation model. Less attention is however given to what
happens in the production process of elevation models.
This thesis project aims to investigate how quality in triangulated elevation models, TINmodels, should be described. Furthermore, what quality affecting factors exist in the production
process and their effects is researched. Lastly, this thesis studies if specifications for data sets
like point clouds and break geometries affect the quality of TIN-models.
The report begins with a literature review to establish central terms and concepts. The research
questions are then studied by conducting an interview study as well as case studies. In the
interview study people that use TIN-models are interviewed about their usage and quality
demands of TIN-models. The interviews indicate that data reduction is an important factor to
optimize production of TIN-models and afterwards have TIN-models small enough to manage.
In the case studies the effects of data reduction, which appeared as an important factor in the
production process, are studied. In addition, the effects of using break geometries are evaluated.
The case studies show that, depending on different factors, large parts of the original data sets
can be reduced. In some cases, less than 1 % of original dataset is needed for creating acceptable
TIN-models. Use of break geometries can also allow for further data reduction if they are well
placed and well produced.
Towards the end of the report the results are discussed and conclusions are drawn. It is discussed
how quality of TIN-models can be described. Additionally, it is discussed how data reduction
and other parameters in source data and production process affect the quality of the final TINmodel.}},
  author       = {{Andersson, Axel}},
  language     = {{swe}},
  note         = {{Student Paper}},
  series       = {{Examensarbete i geografisk informationsteknik}},
  title        = {{Kvalitetsbeskrivning och kvalitetspåverkande faktorer vid produktion av TIN-modeller}},
  year         = {{2024}},
}