Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Rule-based land cover classification and erosion risk assessment of the Krkonose National Park, Czech Republic

H Hyltén, Annika and Uggla, Eva (2000) In Lunds universitets Naturgeografiska institution - Seminarieuppsatser
Dept of Physical Geography and Ecosystem Science
Abstract (Swedish)
Populärvetenskaplig sammanfattning: Krkonoše national park är beläget i Sudeterna vid gränsen mellan Tjeckien och Polen.
Parken karakteriseras av stora variationer i altitud och ett mosaikliknande vegetationsmönster.
Det komplexa landskapet medför svårigheter vid val av klassificeringsteknik
och skala för produktion av tillförlitliga markanvändningskarteringar. Denna studie syftar
i första hand till att producera en markanvändningskarta med 21 klasser baserat på ett
regelbaserat klassificeringssystem samt att utvärdera klassificeringar baserade på denna
metod och på en s.k. ”maximum likelihood” algoritm baserad på Landsat 7 ETM+ data.
Komplement data och sannolikheter baserat på tidigare producerat material används i det
... (More)
Populärvetenskaplig sammanfattning: Krkonoše national park är beläget i Sudeterna vid gränsen mellan Tjeckien och Polen.
Parken karakteriseras av stora variationer i altitud och ett mosaikliknande vegetationsmönster.
Det komplexa landskapet medför svårigheter vid val av klassificeringsteknik
och skala för produktion av tillförlitliga markanvändningskarteringar. Denna studie syftar
i första hand till att producera en markanvändningskarta med 21 klasser baserat på ett
regelbaserat klassificeringssystem samt att utvärdera klassificeringar baserade på denna
metod och på en s.k. ”maximum likelihood” algoritm baserad på Landsat 7 ETM+ data.
Komplement data och sannolikheter baserat på tidigare producerat material används i det
regelbaserade systemet.
Den regelbaserade klassificeringen (21 klasser) uppnår en noggrannhet på 61,5 %. Högre
noggrannhet uppnås om 11 markanvändningsklasser används (74,9 %). Detta indikerar
att resultatet av noggrannhetsutvärderingen för markanvändningskartan med 21 klasser är
starkt påverkad av den relativt låga noggrannheten hos de mindre och mer komplexa
klasserna. Med hänsyn till skillnaderna i altitud, det mosaikliknande vegetationsmönstret
samt det stora antalet klasser är resultatet av det regelbaserade klassificeringssystemet
tillfredställande, speciellt om antalet klasser reduceras. Denna studie visar att ett
regelbaserat system med komplement data otvivelaktigt förbättrar en ”maximum
likelihood” klassificering baserad enbart på spektral data samt att ”maximum likelihood”
klassificeringen ej är tillräcklig för detta område. För att uppnå en bättre klassificering
framstår regelbaserade system med komplement data som en lovande metod för områden
av detta slag.
Det andra syftet med studien är att uppskatta risken för erosion i national parken. Under
de senaste decennierna har höga halter av luftföroreningar deponerats i denna region.
Detta har medfört markförsurning, minskad biodiversitet, omfattande avverkning och
jorderosion. Risken för erosion varierar inom parken, därav är det viktigt att uppskatta
vilka områden som är i farozonen för att bli eroderade. Detta för att motverka reell
erosion. Två klasser används i erosions riskuppskattningen; erosions- kontra icke
erosions riskområden. Modellen baseras på statistiska analyser av fältdata, bestående av
GPS punkter med information om markanvändning och förekomst av erosion, digital data
över jordart, topografisk form, sluttningsgradient, sluttningsriktning och altitud, samt
litteraturstudier. Noggrannhetsutvärderingen gav en noggrannhet på 86,4 %. Detta
indikerar att uppskattningar av denna typ kan utföras med tillfredställande resultat. (Less)
Abstract
The Krkonoše National Park is located in the Sudetes mountain range at the Czech -
Polish border. It is characterized by large variations in altitude and a mosaic vegetation
pattern. The complex landscape raises the issues of appropriate classification techniques
and scales. The first aim of this study is to produce a land cover map using 21 classes
based on a rule-based classification system and to evaluate classifications based on this
method and a maximum likelihood algorithm based on Landsat 7 ETM+ data. Ancillary
data and prior probabilities are used in the rule-based system.
The rule-based classification (21 classes) yields an overall accuracy of 61.5%. A higher
accuracy is reached if 11 land cover classes are used (overall... (More)
The Krkonoše National Park is located in the Sudetes mountain range at the Czech -
Polish border. It is characterized by large variations in altitude and a mosaic vegetation
pattern. The complex landscape raises the issues of appropriate classification techniques
and scales. The first aim of this study is to produce a land cover map using 21 classes
based on a rule-based classification system and to evaluate classifications based on this
method and a maximum likelihood algorithm based on Landsat 7 ETM+ data. Ancillary
data and prior probabilities are used in the rule-based system.
The rule-based classification (21 classes) yields an overall accuracy of 61.5%. A higher
accuracy is reached if 11 land cover classes are used (overall accuracy: 74.9%). This
indicates that the result of the accuracy assessment of the land cover map with 21 classes
is strongly influenced by the rather low accuracy of the more infrequent and complex
classes. Considering the differences in altitude, the mosaic vegetation and the large
number of classes the result of the rule-based classification system is satisfactory,
especially when the number of classes is reduced. This study shows that a rule-based
classification system using ancillary data and prior probabilities clearly enhances a
maximum likelihood classification based solely on spectral data. An interpretation of
satellite data based exclusively on spectral information does not produce a satisfactory
result for this region. To achieve an improved classification the use of ancillary data and
prior probabilities in a rule-based classification system seem to offer a promising
solution.
The second aim of this study is to assess the erosion risk in the National Park. Heavy air
pollution has been deposited in this region during the last decades causing soil
acidification, decreased biodiversity, deforestation and soil erosion. The erosion risk
varies within the park and it is therefore essential to make an assessment of which areas
are in danger of becoming eroded to prevent actual erosion. Two classes are used in the
erosion risk assessment; erosion versus no erosion risk areas. It is based on statistical
analyses of field data, consisting of GPS points including information on land cover and
the presence of erosion /no- erosion, digital data on soil type, topographical form, slope
gradient, aspect and altitude, and on literature studies. Accuracy assessments yield an
overall accuracy of 86.4%. This indicates that assessments of this type can be made with
satisfactory results. (Less)
Please use this url to cite or link to this publication:
author
H Hyltén, Annika and Uggla, Eva
supervisor
organization
alternative title
Regelbaserad markanvändningsklassificering och erosionsuppskattning i Krknose nationalpark, Tjeckien
year
type
H1 - Master's Degree (One Year)
subject
keywords
geomorphology, physical geography, pollution, Czech Republic, mosaic vegetation patterns, soil acidification, mapping, pedology, cartography, climatology, naturgeografi, geomorfologi, marklära, kartografi, klimatologi
publication/series
Lunds universitets Naturgeografiska institution - Seminarieuppsatser
report number
71
language
English
id
1332861
date added to LUP
2005-10-27 00:00:00
date last changed
2011-11-30 12:53:30
@misc{1332861,
  abstract     = {{The Krkonoše National Park is located in the Sudetes mountain range at the Czech -
Polish border. It is characterized by large variations in altitude and a mosaic vegetation
pattern. The complex landscape raises the issues of appropriate classification techniques
and scales. The first aim of this study is to produce a land cover map using 21 classes
based on a rule-based classification system and to evaluate classifications based on this
method and a maximum likelihood algorithm based on Landsat 7 ETM+ data. Ancillary
data and prior probabilities are used in the rule-based system.
The rule-based classification (21 classes) yields an overall accuracy of 61.5%. A higher
accuracy is reached if 11 land cover classes are used (overall accuracy: 74.9%). This
indicates that the result of the accuracy assessment of the land cover map with 21 classes
is strongly influenced by the rather low accuracy of the more infrequent and complex
classes. Considering the differences in altitude, the mosaic vegetation and the large
number of classes the result of the rule-based classification system is satisfactory,
especially when the number of classes is reduced. This study shows that a rule-based
classification system using ancillary data and prior probabilities clearly enhances a
maximum likelihood classification based solely on spectral data. An interpretation of
satellite data based exclusively on spectral information does not produce a satisfactory
result for this region. To achieve an improved classification the use of ancillary data and
prior probabilities in a rule-based classification system seem to offer a promising
solution.
The second aim of this study is to assess the erosion risk in the National Park. Heavy air
pollution has been deposited in this region during the last decades causing soil
acidification, decreased biodiversity, deforestation and soil erosion. The erosion risk
varies within the park and it is therefore essential to make an assessment of which areas
are in danger of becoming eroded to prevent actual erosion. Two classes are used in the
erosion risk assessment; erosion versus no erosion risk areas. It is based on statistical
analyses of field data, consisting of GPS points including information on land cover and
the presence of erosion /no- erosion, digital data on soil type, topographical form, slope
gradient, aspect and altitude, and on literature studies. Accuracy assessments yield an
overall accuracy of 86.4%. This indicates that assessments of this type can be made with
satisfactory results.}},
  author       = {{H Hyltén, Annika and Uggla, Eva}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Lunds universitets Naturgeografiska institution - Seminarieuppsatser}},
  title        = {{Rule-based land cover classification and erosion risk assessment of the Krkonose National Park, Czech Republic}},
  year         = {{2000}},
}