Latest advance in remote sensing technologies have introduced
new and efficient methods to acquire information about the Earth
such as the Ikonos satellite imaging system and the Light
Detection and Ranging (LIDAR) system. While Ikonos imagery
measures the spectral reflectance of the ground, operating at the
visible range of the electromagnetic spectrum, LIDAR data are
geometric range measurements, operating with a wavelength of
about 1000 eÀ6 m. When the wave hits the Earth it is reflected
back to the sensor and the time it took to transmit and return is
recorded. The returned signal can then be processed to extract
information about the Earth. Therefore, the combination of these
two measurements provides accurate geometric and spectral
information about the ground, which could be used to produce
high-quality topographic maps, Ackermann [2].
This research shows the potential of utilizing a 1-m resolution
LIDAR-based Digital Elevation Model (DEM) to orthorectify a single Ikonos Geo-panchromatic image. The planimetric accuracy of the LIDAR-based DEM is first evaluated using 12 Ground Control
Points (GCPs) surveyed using the Differential GPS (DGPS)
technique. Two shifts of about 6 to 8 m in the East and North
directions are found. These shifts are removed and the LIDAR-
based DEM is then used to collect new control points and
checkpoints. These points are used in the rectification and
orthorectification process of the Ikonos imagery using different
mathematical models with different sets of GCPs and checkpoints.
The affine and projective 2-D transformation models showed a
Root Mean Square Errors (RMSE) of about 5 m in the rectification
process of the Ikonos image.
The relationship between the planimetric errors and the
elevations of the checkpoints is examined and it suggested using
3-D to 2-D transformation models to orthorectify the Ikonos
image in order to provide better accuracy. Based on these results,
three different 3-D to 2-D transformation models are used. First, a
1st order and a 2nd order relief-corrected transformation models
are used to orthorectify the Ikonos image. Results show that about
1.7–1.4 m horizontal RMSE could be achieved using 14 GCPs. In
addition, the DLT transformation model is also tested and it
showed approximately about 1.30 m horizontal RMSE. These
results show that the required planimetric accuracy for 1:5000
topographic maps could be achieved using a 1-m resolution
LIDAR-based DEM and a single Ikonos panchromatic image. As a
final product, the 1st order relief-corrected transformation model
is used to generate the orthorectified image from the Ikonos
image and the LIDAR-based DEM.
The paper is organized in the following order. An overview of
current studies on the geometric accuracy of Ikonos imagery is
first presented. The next section presents a brief description of the
used datasets. Then the rectification process of the LIDAR-based
DEM is discussed. The next section describes the use of 2-D
transformation models to rectify the Ikonos imagery. Then the
process and results of orthorectifying the Ikonos image are
described and analyzed. In the last section conclusions and future
work are discussed.