Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Norwegian Open Resea...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
Brage NMBU
Master thesis . 2019
Data sources: Brage NMBU
versions View all 2 versions
addClaim

Endringsdeteksjon i skog med Sentinel-2

Authors: Sletmo, Jonas Njøten;

Endringsdeteksjon i skog med Sentinel-2

Abstract

Sentinel-2 har vært operativt siden 2015 og har i perioden etter det blitt benyttet til jordovervåkning, klimaovervåkning og nødadministrasjonstjenester. Sentinel-2 med sin høye romlige og temporale oppløsning er et kraftfullt verktøy som har vært benyttet til å finne endringer på jordoverflaten over tid. Denne oppgaven fokuserer på hvordan man kan bruke informasjon fra Sentinel-2 på best mulig måte for å finne endringer i skog (hogster). Jeg har benyttet to Sentinel-2 bilder, det første fra 2017 og det andre fra 2018. Disse bildene har jeg hentet fra Copernicus Open Access Hub. I oppgaven benyttes en nevral nettverksmodell til å finne de nevnte endringer. Det er benyttet tre datasett som er satt opp forskjellige måter. Det ene inneholder data fra 2018 og 2017. Det neste inneholder data fra 2018 og differansene mellom 2018 og 2017. Det tredje kun differanser mellom 2018 og 2017. For de tre datasettene har jeg sett på forskjellige kombinasjoner av spektralbånd. Resultatene av mine tester viser at både valg av spektralbånd og oppsett av datasett har en innvirkning på hvor gode resultatene blir. Mengden av treningsdata har stor innvirkning på hvor godt en modell klarer å predikere resultatet. Resultatene viser også at et godt datagrunnlag er viktig for å oppnå et gode resultater. De beste resultatene fremkommer ved bruk av alle tilgjengelige 10 og 20 meters bånd, i tillegg gir testen med 10 meters bånd bedre resultater enn testen med 20 meters bånd.

Sentinel-2 has been operational since 2015 and has since then been used for land monitoring, climate monitoring and emergency management. Sentinel-2 with its high spatial and temporal resolutions is a powerful tool that has been used to detect changes on the earth's surface over time. This thesis focuses on how to use information from Sentinel-2 in the best possible way to find changes in forest (logging). I have used two Sentinel-2 images, the first was taken in 2017 and the second in 2018. I have obtained these images from the Copernicus Open Access Hub. I have uses a neural network model to find the aforementioned changes. I have used three different datasets that have been set up in different ways. One contains data from 2018 and 2017. The next contains data from 2018 and the differences between 2018 and 2017. The third only difference between 2018 and 2017. For the three datasets I have looked at different combinations of spectral bands. The results of my tests show that both the choice of spectral bands and the set-up of data sets have an impact on how good the results become. The amount of training data has a great impact on how well a model manages to predict the result. The results also show that a good data base is important for achieving good results. The best results are obtained with the use of all available 10- and 20-meter bands, in addition, the test with 10 meters of band gives better results than the test with 20 meters band.

M-GEOM

Country
Norway
Related Organizations
Keywords

Endringsanalyse, VDP::Technology: 500, Geomatikk, Sentinel-2

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green
Related to Research communities