The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part BI. Beijing 2008
369
DTM
5m
(nominal)
Type I: 0.7m
Type II: 1.0m
WGS84/
EGM96/
Geographic
ORI
1.25m or
2.5 m
(nominal)
2.0m
WGS84/
Geographic
Baseline
length (m)
0.93
0.96
0.93
0.94
Look
direction
left or
right
left or
right
left or
right
left or
right
Best image
solution (m)
1.25
1.25/0.6
1.25/0/6
1.25/0.6
Table 1: System Parameters of Intermap’s STAR Systems (*
Terrain dependent)
2.3 Airborne IFSAR Mapping Products
IFSAR mapping is essentially a process of producing 3-D map
products by processing raw radar data collected by airborne
IFSAR systems. Ground thematic information is derived from
the synthetic aperture radar (SAR) images. Height information
is obtained by using the phase difference between two coherent
SAR images simultaneously obtained by two antennas
separated by an across-track baseline in a single-pass mode. The
major steps of the operation and production workflow of a
typical airborne IFSAR mission are summarized as follows:
• Mission planning and acquisition,
• Field QC and Navigation process,
• SAR image formation,
• Interferometric process and geo-coding,
• Editing and post-process.
Figure 2 illustrates the mapping process of the STAR
technology, see Li et al, 2004 and Bryan, 2007 for the details of
the STAR system operation and production process.
The direct products of Intermap’s IFSAR systems are the
Orthorectified Radar Image (ORI) and the Digital Surface
Model (DSM) that represents the surface visible to the radar
sensor. With the Intermap’s proprietary bare-earth processing
methodology a bare-earth Digital Terrain Model (DTM) is
derived for many terrain types in order to deliver the
topographic surface of the Earth. ORI, DSM and DTM are three
core products of Intermap’s airborne IFSAR mapping system.
Table lists the major parameters of three core products.
STAR
Product
Post
Spacing
RMSE
Accuracy
Datum
Coordinate
Systems
DSM
5m
(nominal)
Type I: 0.5m
Type II: 1.0m
Type III: 3.0m
WGS84/
EGM96/
Geographic
Table 2: Specifications of Intermap's Core Products
(* - Other datum, projections and coordinate systems are also
supported depending on the area and requirements.)
2.4 Image Quality and DEM Accuracy
2.4.1 Image Quality: In general the image quality of SAR
image is measured by following three parameters of the system
impulse response:
• Image resolution (mainlobe -3dB width)
• Peak side-lobe ratio (PSLR)
• Integrated side-lobe ratio (ISLR)
Table 3 lists the image quality parameters of STAR systems
with different bandwidth. As indicated in Table 3 Intermap’s
STAR systems can provide the image with spatial range
resolution of 1.25 and 0.65 m for the pulse bandwidth of 135
and 270 MHz.
STAR
system
Image
resolution (m)
PSLR (dB)
ISLR (dB)
135 MHz
1.26-1.3
-18.3-22.3
-35.7-40.2
270 MHz
0.65-0.68
-22.2-24.2
-36.7-38.8
Table 3: Image quality of STAR system
Figure 3 shows the comparison of images with different
resolution (pulse bandwidth) for the same area. As
demonstrated in Figure 3 the edge definition of the roads and
the edge of buildings are noticeably improved by the high
resolution images of STAR systems with 135 MHz band width.
2.4.2 DEM Accuracy: Applying the error propagation to
equations (1) and (2) for geocoding using interferometric
process results the height error, in terms of standard deviation,
can be expressed by
°h =
Oh =
sin{(p - a)cos ß ■ a p
(3)
p cos tp cos(tp -a) cos ß
„ . o b
(4)
a sirup
 pcos(tp- a)cosß
(5)
4 n B sirup
p• cos(rp-a)cosß • <7 a
(6)
p ■ sin(ç? - a) sin ß ■ <7ß
(7)
In equation (3) to (7) the time delay and the bias of the baseline
length can be precisely estimated by the calibration. Since the
pitch error is not sensitive to the height error, the
interferometric phase error and the roll error of the SAR
antenna platform are the dominating error sources of the D T E
heights.
Figure 4 shows the histogram of typical height errors comparea
to the reference DEM with the statistics of DEM errors: mean =
0.012 m and standard deviation = ±0774 m. Figure 5 shows the
height errors compared to external ground control points (GCP)