University of Twente Student Theses
Integrating Landsat ETM+ and DEM to Support Forest Management Map in Part of Papua Province, Indonesia
Ikwan, I. (2010) Integrating Landsat ETM+ and DEM to Support Forest Management Map in Part of Papua Province, Indonesia.
|
PDF
12MB |
| Abstract: | There is a big challenge in the integration between land to conserve and land to develop in regional planning. In some areas with the lack of high spatial resolution information, due to the scale problem, landscape approach using remote sensing and GIS can contribute as an input in regional spatial planning. The aim of this study was to test the capabilities of Neural Network Classification, using Landsat ETM+ and DEM, to produce useful information with sufficient accuracy. Spatial planning maps of Jayapura City and Cycloop Nature Reserve, were used to define the gap of management that emerge. ASTER GDEM and SRTM DEM compared to select the appropriate DEM sources to derived morphometric elements in the study area. Several combinations of Neural Network Classifications were done to combine Landsat ETM+ and DEM for deriving the morphometric element in the study area. Further, the study compared Maximum Likelihood and Neural Network Classification. Thus, the study finished with detailing land cover information in the gap area. Using cross section (profile) and minimum curvature, the study shows that SRTM DEM give more accurate information of elevation. Neural Network classifier using Landsat ETM+ and extra information of elevation and classes of elevation can improve the overall accuracy by 10 % compared to standard statistical classifier (i.e. maximum likelihood). Elevation information from ASTER GDEM were unreliable because of the stacking processes in ASTER GDEM. To further improve the classification results, Neural Network classifier need more extra information which are closely related with the land cover types to classify them. |
| Item Type: | Essay (Master) |
| Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
| Programme: | Geoinformation Science and Earth Observation MSc (75014) |
| Link to this item: | https://purl.utwente.nl/essays/90715 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page