hig.sePublikationer
Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • harvard-cite-them-right
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • sv-SE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • de-DE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Data source evaluation for shoreline deliniation applications
Högskolan i Gävle, Akademin för teknik och miljö, Avdelningen för Industriell utveckling, IT och Samhällsbyggnad, Datavetenskap. (Datavetenskap)
Högskolan i Gävle, Akademin för teknik och miljö, Avdelningen för Industriell utveckling, IT och Samhällsbyggnad, Datavetenskap. Uppsala University, Department of Information Technology, Sweden. (Datavetenskap)ORCID-id: 0000-0003-0085-5829
Technische Universität, Dresden, Germany.
2017 (Engelska)Ingår i: International Multidisciplinary Scientific GeoConference Surveying Geology and Mining Ecology Management, SGEM: Conference proceedings, 2017, Vol. 17, nr 2-3, s. 849-858Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

This paper proposes an evaluation of data acquired with various sensors and used in coastal water segmentation applications. Correct monitoring of coastal changes in dynamic coastal environments strongly depends on accurate and frequent detection of shoreline position. Automatic shoreline delineation methods are preferable, especially in terms of time, cost, labor intensiveness and difficulties of in-situ measurements. Two main issues have been encountered within this application field, the quality of data and the segmentation algorithms. In this work, potential benefits of various data sources including optical and active sensors for extraction of shorelines have been investigated. The goal with shoreline detection from digital data sources is to obtain information as efficiently as possible and as reliably as necessary. Starting with that observation the paper discusses the effectiveness of coastal information extraction provided different data sources. This question is especially important to address since we observe a fast development of high spatial resolution data acquisition. There are many of segmentation algorithms described in the field of image processing and yet there is currently no single theory or method, no universal segmentation framework, that can be applied on all images to precisely and robustly extract shorelines. Nether there is a uniform standard for the assessment of segmentation results, and this process still largely relies on visual analysis and personal judgment. Out of myriads of image segmentation algorithms, we chose the most frequently and successfully applied within the application field and considering the data sources. In optical sensor data cases, the most frequently used methods are NDWI (Normalized Difference Water Index) and thresholding techniques. We do not aim to create yet another method to segment out the particular objects from remotely sensed data and then tailor it to work efficiently on that data set. Instead, we evaluate the data quality regarding the given application field. The case study is carried out on a 10 km coastal stretch facing the Baltic Sea (Sweden) and belonging to the Municipality of Gävle. In citu measurements were acquired to evaluate the extracted coastal lines and comparisons with reference were performed based on the average mean distance. A conclusion is done regarding the most reliable data source for this particular application of shoreline delineation.

Ort, förlag, år, upplaga, sidor
2017. Vol. 17, nr 2-3, s. 849-858
Serie
International Multidisciplinary Scientific GeoConference Surveying Geology and Mining Ecology Management, SGEM : Conference proceedings, ISSN 1314-2704 ; 21
Nyckelord [en]
Data source, Evaluation, Segmentation, Shoreline
Nationell ämneskategori
Data- och informationsvetenskap Geovetenskap och miljövetenskap
Identifikatorer
URN: urn:nbn:se:hig:diva-25424DOI: 10.5593/sgem2017/21/S08.108Scopus ID: 2-s2.0-85032471686ISBN: 978-619-7408-01-0 (tryckt)OAI: oai:DiVA.org:hig-25424DiVA, id: diva2:1150210
Konferens
17th International Multidisciplinary Scientific GeoConference SGEM 2017,27 June - 6 July, 2017, Albena, Bulgaria
Tillgänglig från: 2017-10-18 Skapad: 2017-10-18 Senast uppdaterad: 2018-03-13Bibliografiskt granskad

Open Access i DiVA

Fulltext saknas i DiVA

Övriga länkar

Förlagets fulltextScopus

Personposter BETA

Åhlén, JuliaSeipel, Stefan

Sök vidare i DiVA

Av författaren/redaktören
Åhlén, JuliaSeipel, Stefan
Av organisationen
Datavetenskap
Data- och informationsvetenskapGeovetenskap och miljövetenskap

Sök vidare utanför DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetricpoäng

doi
isbn
urn-nbn
Totalt: 72 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • harvard-cite-them-right
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • sv-SE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • de-DE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf