Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • 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
Detecting ditches using supervised learning on high-resolution digital elevation models
Jönköping University, SWE.
Jönköping University, SWE.
Blekinge Institute of Technology, Faculty of Computing, Department of Software Engineering.ORCID iD: 0000-0002-0535-1761
Swedish Forest Agency, SWE.
Show others and affiliations
2022 (English)In: Expert systems with applications, ISSN 0957-4174, E-ISSN 1873-6793, Vol. 201, article id 116961Article in journal (Refereed) Published
Abstract [en]

Drained wetlands can constitute a large source of greenhouse gas emissions, but the drainage networks in these wetlands are largely unmapped, and better maps are needed to aid in forest production and to better understand the climate consequences. We develop a method for detecting ditches in high resolution digital elevation models derived from LiDAR scans. Thresholding methods using digital terrain indices can be used to detect ditches. However, a single threshold generally does not capture the variability in the landscape, and generates many false positives and negatives. We hypothesise that, by combining the digital terrain indices using supervised learning, we can improve ditch detection at a landscape-scale. In addition to digital terrain indices, additional features are generated by transforming the data to include neighbouring cells for better ditch predictions. A Random Forests classifier is used to locate the ditches, and its probability output is processed to remove noise, and binarised to produce the final ditch prediction. The confidence interval for the Cohen's Kappa index ranges [0.655, 0.781] between the evaluation plots with a confidence level of 95%. The study demonstrates that combining information from a suite of digital terrain indices using machine learning provides an effective technique for automatic ditch detection at a landscape-scale, aiding in both practical forest management and in combatting climate change. © 2022 The Authors

Place, publisher, year, edition, pages
Elsevier Ltd , 2022. Vol. 201, article id 116961
Keywords [en]
Classification and regression trees, Geographic information systems, Machine learning, Supervised learning by classification, Classification (of information), Climate change, Decision trees, Digital instruments, E-learning, Forestry, Gas emissions, Geomorphology, Greenhouse gases, Information use, Metadata, Supervised learning, Surveying, Wetlands, Classification trees, Digital elevation model, Digital terrain, Drainage networks, Forest production, Greenhouse gas emissions, High resolution, Landscape scale, Regression trees
National Category
Physical Geography Remote Sensing
Identifiers
URN: urn:nbn:se:bth-22881DOI: 10.1016/j.eswa.2022.116961ISI: 000830107400002Scopus ID: 2-s2.0-85128240716OAI: oai:DiVA.org:bth-22881DiVA, id: diva2:1654976
Funder
VinnovaSwedish Research Council Formas
Note

open access

Available from: 2022-04-29 Created: 2022-04-29 Last updated: 2022-08-12Bibliographically approved

Open Access in DiVA

fulltext(6026 kB)341 downloads
File information
File name FULLTEXT01.pdfFile size 6026 kBChecksum SHA-512
49b191053be4df971a23993af724b1936f2c9c1ecd6f5a654a2a1971c9b8407fa908f4013d7914eadbaac0c3f50b0466a5541cdb9d4c012ad1c397afa889e8de
Type fulltextMimetype application/pdf

Other links

Publisher's full textScopus

Authority records

Lavesson, Niklas

Search in DiVA

By author/editor
Lavesson, Niklas
By organisation
Department of Software Engineering
In the same journal
Expert systems with applications
Physical GeographyRemote Sensing

Search outside of DiVA

GoogleGoogle Scholar
Total: 341 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 181 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • 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