Advanced

The potential of support vector machine classification of land use and land cover using seasonality from MODIS satellite data

Sallaba, Florian (2011) In Lunds universitets Naturgeografiska institution - Seminarieuppsatser
Dept of Physical Geography and Ecosystem Science
Abstract
With respect to climate change it is necessary to study land use and land cover (LULC) and
their changes. LULC are related directly and indirectly to climatic changes such as rising
temperatures that trigger earlier onset of vegetation growing seasons (IPCC 2007). Land
surface phenology refers to the seasonal patterns of variation in vegetated land surfaces over
large areas using satellite data (Reed et al. 2009). General variations observed from satellite
may also be referred to as seasonality (Jönsson and Eklundh 2002, 2004).
In this study, seasonality was modeled from normalized difference vegetation index timeseries
derived from Moderate Resolution Imaging Spectro-Radiometer (MODIS) satellite
data. Seasonality data contain... (More)
With respect to climate change it is necessary to study land use and land cover (LULC) and
their changes. LULC are related directly and indirectly to climatic changes such as rising
temperatures that trigger earlier onset of vegetation growing seasons (IPCC 2007). Land
surface phenology refers to the seasonal patterns of variation in vegetated land surfaces over
large areas using satellite data (Reed et al. 2009). General variations observed from satellite
may also be referred to as seasonality (Jönsson and Eklundh 2002, 2004).
In this study, seasonality was modeled from normalized difference vegetation index timeseries
derived from Moderate Resolution Imaging Spectro-Radiometer (MODIS) satellite
data. Seasonality data contain valuable information about vegetation dynamics of LULC,
such as the maximum of a season as well as the season start and end. The specific seasonality
data signatures of LULC and may improve LULC classifications compared to multi-spectral
satellite data approaches.
Support vector machine classification (SVC) is a machine learning technique that does not
require normal distributed input data. A normal distribution of seasonality data cannot be
assumed. SVC is superior in comparison to traditional classification methods using multispectral
satellite data (Tso and Mather 2009). Thus, it is feasible to test the potential of SVC
separation of LULC using seasonality data. The most common linear and non-linear SVC
methods recommended for satellite data were applied in this study.
The chosen study area is located in southern Sweden, and its LULC classes are well
documented by the latest CORINE land cover 2006 data. Thus, it is a good test area for
validation of the performance of seasonality parameters for LULC classification using SVC.
In this study, a SVC framework was developed and implemented that: (1) selects the most
appropriate input seasonality data, (2) incorporates a direct acyclic graph for multiclassification
and (3) validates the SVC outcomes with an accuracy assessment.
The results of the four class SVC show moderate performances with overall accuracies
between 61 - 64% and Kappa values ranging from 0.42 – 0.45. The accuracy differences
between linear and non-linear SVC are marginal. However, there are potentials to improve
the developed methodology, and thus the performance of SVC on seasonality data. In
addition, the seasonality data should be tested with traditional parametric classifiers (i.e.
maximum likelihood) in order to achieve valuable comparisons. (Less)
Abstract (Swedish)
Då klimatförändringar studeras är det viktigt att ha markanvändningarna och deras
förändringar i åtanke. Markanvändningarna är direkt och indirekt relaterat till
klimatförändringar såsom exempelvis stigande temperaturer, vilket påskyndar starten av
växtsäsongen. Fjärranalysdata ger en bättre bild av storskaliga förändringar i vegetationen än
vad som är möjligt att observera från jordytan. Vegetationen och dess fenologi kan
observeras med satellitdata, och kallas även för årstidsvariationer.
Denna studie använder tidsserier av vegetationsindex från satellitdata för att modellera
årstidsvariationer. Dessa matematiska modeller av årstidsvariationerna används sedan för att
extrahera olika årstidsrelaterade parametrar som ger värdefull... (More)
Då klimatförändringar studeras är det viktigt att ha markanvändningarna och deras
förändringar i åtanke. Markanvändningarna är direkt och indirekt relaterat till
klimatförändringar såsom exempelvis stigande temperaturer, vilket påskyndar starten av
växtsäsongen. Fjärranalysdata ger en bättre bild av storskaliga förändringar i vegetationen än
vad som är möjligt att observera från jordytan. Vegetationen och dess fenologi kan
observeras med satellitdata, och kallas även för årstidsvariationer.
Denna studie använder tidsserier av vegetationsindex från satellitdata för att modellera
årstidsvariationer. Dessa matematiska modeller av årstidsvariationerna används sedan för att
extrahera olika årstidsrelaterade parametrar som ger värdefull information om
vegetationsdynamiken. Dessa årstidsrelaterade parametrar kan förbättra klassificeringen av
markanvändningarna.
Studien tillämpar en ny teknik, som kallas ”support vector machine” klassificering, för att
klassificera de årstidsrelaterade parametrarna. Studien fokuserade på de vanligaste linjära och
olinjära ”support vector machine” tekniker som rekommenderas för satellitdata.
Det studerade området ligger i södra Sverige och dess markanvändningsklasser är sedan
tidigare väldokumenterade. Det innebär att området är mycket lämpligt för att testa
årstidsrelaterade parametrar genom att tillämpa ”support vector machine” klassificering.
Studien utgick från att: (1) välja de mest lämpliga årstidsrelaterade parametrarna, (2) använda
en multi-klassificering, och (3) utvärdera de klassificeringsutfall med
noggrannhetsbedömningar baserat från senast dokumenterad CORINE land cover 2006 data.
Resultaten påvisar en måttlig prestanda hos ”support vector machine” klassificering, där den
övergripande noggrannheten landar mellan 61 till 64 %, och Kappavärdet varierar mellan
0,49 till 0,52. Skillnaderna mellan de linjära och olinjära ”support vector machine”
teknikerna är marginella. Dock bör årstidsrelaterade parametrar klassificeras med
traditionella klassificeringsmetoder (t.ex. maximum likelihood-metoden), och jämföras med
”support vector machine” klassificeringsresultaten. (Less)
Please use this url to cite or link to this publication:
author
Sallaba, Florian
supervisor
organization
year
type
H2 - Master's Degree (Two Years)
subject
keywords
geography, geomatics, remote sensing, MODIS, support vector machine, phenology, TIMESAT, geografi, geomatik, fjärranalys, fenologi
publication/series
Lunds universitets Naturgeografiska institution - Seminarieuppsatser
report number
220
language
English
id
2158702
date added to LUP
2012-03-20 10:37:45
date last changed
2012-03-20 10:37:45
@misc{2158702,
  abstract     = {With respect to climate change it is necessary to study land use and land cover (LULC) and
their changes. LULC are related directly and indirectly to climatic changes such as rising
temperatures that trigger earlier onset of vegetation growing seasons (IPCC 2007). Land
surface phenology refers to the seasonal patterns of variation in vegetated land surfaces over
large areas using satellite data (Reed et al. 2009). General variations observed from satellite
may also be referred to as seasonality (Jönsson and Eklundh 2002, 2004).
In this study, seasonality was modeled from normalized difference vegetation index timeseries
derived from Moderate Resolution Imaging Spectro-Radiometer (MODIS) satellite
data. Seasonality data contain valuable information about vegetation dynamics of LULC,
such as the maximum of a season as well as the season start and end. The specific seasonality
data signatures of LULC and may improve LULC classifications compared to multi-spectral
satellite data approaches.
Support vector machine classification (SVC) is a machine learning technique that does not
require normal distributed input data. A normal distribution of seasonality data cannot be
assumed. SVC is superior in comparison to traditional classification methods using multispectral
satellite data (Tso and Mather 2009). Thus, it is feasible to test the potential of SVC
separation of LULC using seasonality data. The most common linear and non-linear SVC
methods recommended for satellite data were applied in this study.
The chosen study area is located in southern Sweden, and its LULC classes are well
documented by the latest CORINE land cover 2006 data. Thus, it is a good test area for
validation of the performance of seasonality parameters for LULC classification using SVC.
In this study, a SVC framework was developed and implemented that: (1) selects the most
appropriate input seasonality data, (2) incorporates a direct acyclic graph for multiclassification
and (3) validates the SVC outcomes with an accuracy assessment.
The results of the four class SVC show moderate performances with overall accuracies
between 61 - 64% and Kappa values ranging from 0.42 – 0.45. The accuracy differences
between linear and non-linear SVC are marginal. However, there are potentials to improve
the developed methodology, and thus the performance of SVC on seasonality data. In
addition, the seasonality data should be tested with traditional parametric classifiers (i.e.
maximum likelihood) in order to achieve valuable comparisons.},
  author       = {Sallaba, Florian},
  keyword      = {geography,geomatics,remote sensing,MODIS,support vector machine,phenology,TIMESAT,geografi,geomatik,fjärranalys,fenologi},
  language     = {eng},
  note         = {Student Paper},
  series       = {Lunds universitets Naturgeografiska institution - Seminarieuppsatser},
  title        = {The potential of support vector machine classification of land use and land cover using seasonality from MODIS satellite data},
  year         = {2011},
}