uu.seUppsala University Publications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Automatic Water Body Extraction From Remote Sensing Images Using Entropy
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Visual Information and Interaction. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Computerized Image Analysis and Human-Computer Interaction.ORCID iD: 0000-0003-0085-5829
2015 (English)In: SGEM2015 Conference Proceedings, 2015, Vol. 2, 517-524 p.Conference paper, Published paper (Refereed)
Abstract [en]

This research focuses on automatic extraction of river banks and other inland waters from remote sensing images. There are no up to date accessible databases of rivers and most of other waters objects for modelling purposes. The main reason for that is that some regions are hard to access with the traditional ground through techniques and thus the boundary of river banks are uncertain in many geographical positions. The other reason is the limitations of widely applied method for extraction of water bodies called normalized-difference water index (NDWI). There is a novel approach to extract water bodies, which is based on pixel level variability or entropy, however, the methods work somewhat satisfactory on high spatial resolution images, there is no verification of the method performance on moderate or low resolution images. Problems encounter identification of mixed water pixels and e.g. roads, which are built in attachment to river banks and thus can be classified as rivers. In this work we propose an automatic extraction of river banks using image entropy, combined with NDWI identification. In this study only moderate spatial resolution Landsat TM are tested. Areas of interest include both major river banks and inland lakes. Calculating entropy on such poor spatial resolution images will lead to misinterpretation of water bodies, which all exhibits the same small variation of pixel values as e.g. some open or urban areas. Image entropy thus is calculated with the modification that involves the incorporation of local normalization index or variability coefficient. NDWI will produce an image where clear water exhibits large difference comparing to other land features. We are presenting an algorithm that uses an NDWI prior to entropy processing, so that bands used to calculate it, are chosen in clear connection to water body features that are clearly discernible.As a result we visualize a clear segmentation of the water bodies from the remote sensing images and verify the coordinates with a given geographic reference.

Place, publisher, year, edition, pages
2015. Vol. 2, 517-524 p.
Series
International Multidisciplinary Scientific GeoConference-SGEM, ISSN 1314-2704
Keyword [en]
water; NDWI; Entropy; extraction
National Category
Computer Vision and Robotics (Autonomous Systems)
Research subject
Computerized Image Processing
Identifiers
URN: urn:nbn:se:uu:diva-266903DOI: 10.5593/SGEM2015/B21/S8.064ISI: 000371599500064ISBN: 9786197105346 (print)OAI: oai:DiVA.org:uu-266903DiVA: diva2:869180
Conference
15th International Multidisciplinary Scientific GeoConference SGEM 2015
Available from: 2015-09-15 Created: 2015-11-13 Last updated: 2018-01-10Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full text

Authority records BETA

Seipel, Stefan

Search in DiVA

By author/editor
Seipel, Stefan
By organisation
Division of Visual Information and InteractionComputerized Image Analysis and Human-Computer Interaction
Computer Vision and Robotics (Autonomous Systems)

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 487 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf