| Summary: | The rapid development in an area, required more geospatial information. One
alternative for obtaining geospatial information with high precision, wide area
coverage and rapid data collection is by using orthorectified image and rectified
image. To make orthorectified image, it needs Digital Elevation Model (DEM) data
and ground control points (GCPs), while to make rectified image need GCPs.
However, how various DEM and the number of GCP affect the quality of
orthorectified image currently is not exactly defined. How various the number of
GCP affect the quality of rectified image also currently is not exactly defined. This
study aimed to evaluate the effects of various DEM quality and various number of
GCP to the result of orthorectified image and also the effects of various number of
GCP to the result of rectified image.
This research employed the following data: two satellite imageries with spatial
resolution 60 cm on Kaliurang area with slope more than 20% and on UGM with
slope less than 2%, four DEM namely LiDAR DEM with accuracy 30 cm, XSAR
DEM with accuracy 5 m, SRTM 90m DEM with accuracy 16 m, and ASTER GDEM
with accuracy 17 m, and also orthophoto with map scale 1 : 1000. The width of the
study area is 4 km
2
. The research started with determining of GCP and check points
(CP) on the orthophoto map and satellite imagery as much as 20 GCPs and 20 CPs
on the Kaliurang area as well as 20 GCPs and 20 CPs on UGM area.
Orthorectification with Rational Polynomial Coefficients (RPC) method was
performed by using various DEM quality and various number of GCP, while
rectification with second order polynomial method was performed by using various
number of GCP. The coordinate of check points were identified on the orthorectified
image, on the rectified image, and on the orthophoto map in the Kaliurang area and
in the UGM area. Check points coordinate observed on the orthophoto map were
compared to those observed on the orthorectified image and on the rectified image.
Statistical test was implemented on the check point orthorectified image and on the
rectified image in the Kaliurang area and in the UGM area. Root Mean Square Error
(RMSE) then was computed and compared to the BIG accuracy standard.
The result shows that the number of GCP have more effect to the accuracy of
orthorectified image than the quality of DEM. The orthorectified image have an
accuracy of less than 1.5 m, while the rectified image have an accuracy of more than
1,5 m. According to BIG accuracy standard, orthorectified image and rectified image
can be used to make a map of scale at 1 : 2500 and 1 : 5000. Base on the scale,
orthorectified image and rectified image from this research can be used for spatial
planning, base map for land registration, and for village boundary making for the
village.
|
|---|