DEVELOPMENT OF PREANALYSIS MODEL FOR PREDICTION OF ACCURACY IN SATELLITE DERIVED BATHYMETRY (SDB) DUE TO THE EFFECT OF DIFFERENT IMAGE SPATIAL RESOLUTION
Satellite Derived Bathymetry (SDB) is a way to obtain depth data by utilizing remote sensing. This method is considered efficient, but has not been widely used because many parameters can affect the accuracy value. Parameters that affect accuracy such as electromagnetic wave propagation medium, atmo...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/67306 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Satellite Derived Bathymetry (SDB) is a way to obtain depth data by utilizing remote sensing. This method is considered efficient, but has not been widely used because many parameters can affect the accuracy value. Parameters that affect accuracy such as electromagnetic wave propagation medium, atmospheric conditions, physical properties of water, seabed and others, it is necessary to do research to determine the parameters that affect the level of accuracy. Understanding the parameters that can affect the accuracy value can be used to predict the accuracy value that will be generated. The purpose of this study is to create a preanalytical model that predicts the value of accuracy.
This preanalysis model was created by utilizing three images with different spatial resolutions, namely, Landsat 5, Sentinel 2 and PlanetScope 3. The depth extraction method used is the band ratio method. The reason for choosing this method is because it uses two channels with different wavelengths. This band ratio method assumes that two channels with different wavelengths can give different spectral responses to water objects. The channel used is a combination of the blue channel divided by the green channel. This combination was chosen because when regression was performed it produced the best coefficient of determination (R²). Based on the preanalysis model, Landsat 5 imagery can produce an accuracy value of 1.288 meters. Sentinel 2 image produces an accuracy value of 0.627 meters. PlanetScope image with an accuracy value of 0.613 meters. So it can be concluded that the better the image resolution used, the better the accuracy value produced. This is because the better the spatial resolution, the more pure pixel content it has.
|
|---|