Combined
Point/Line
RMS | Erdas I/RAS | Point- Line-
magine C based | based
X m) 20.06 19.87 20.06 19.81 19.96
Y(m) 20.73 20.29 20.71 21.04 20.80
Table 7. RMS Values of 25 Check Points for SPOT4
Multispectral Image Using Point, Line, and Combined Point/Line
Techniques
3.4 IRS-1D
An IRS-ID (Indian Remote Sensing satellite) panchromatic
image with 5.8 m resolution was used in this experiment. The
image covered the Toshka area in the southern west part of Egypt
(Figure 5).
anm
Figure 5. IRS-1D Panchromatic Image over Toshka with its
Control and Check Points Distribution
The developed program for the point-based projective equations
was applied first with different number of control points. The
best possible accuracy of about 1.4 pixels was achieved using 27
control points. RMS values of 25 checkpoints are presented in
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004
RMS i Point- Line- Combined
[RASC based based Point/Line
X (m) 8.36 8.43 8.30 8.35
Y (m) 7.55 7.50 7.87 7.66
Table 9. RMS Values of 25 Check Points for IRS-1D
Panchromatic Image Using Point, Line, and Combined
Point/Line Techniques
3.5 IKONOS
The final satellite imagery experiment was performed on the
high-resolution satellite; IKONOS. A panchromatic IKONOS
image, of 1-m resolution, over the East Cairo region, was used
in this experiment (Figure 6).
Figure 6. IKONOS Panchromatic Image over East Cairo with
its Control (blue) and Check (white) Features Distribution
The developed program for the point-based projective
equations was used to rectify the image using different number
of control points. A number of 27 control points were the
minimum required to achieve the best possible accuracy of
about 1.7 pixels for this type of high-resolution image. RMS
values of 25 checkpoints are presented in Table 10.
Table 8. RMS Values of 25 Check Points for IRS-1D
Panchromatic Image Using Point-based Technique with Various
Number of Control Points
Next, the results of the developed program for point-based
projective equations were compared with those obtained from
I/RASC using 27 control points, as shown in Table 9, to confirm
the accuracy of the developed program. Then, line-based and
combined point/line-based developed programs were applied to
check the required minimum number of control features to obtain
the best possible accuracy figures. As shown in Table 9, only 5
control lines in the line-based projective equations and 3 control
lines & 2 control points in the combined point/line projective
equations, were needed to achieve the equivalent RMS values of
the same 25 check points.
Table 8. RMS s TO control 1^ 20 27
control x control | control | control
RMS 5 10 i5 20 57 points points points points points
coniroil- controi control control. | control X(m) | 32.22 20.65 15.65 8.62 1.79
points | points points points | points Y (m) | 32.58 21.00 15.00 8.45 1.65
X (m) | 94.65 64.87 45.95 28.64 8.43
Y (m) 95.78 64.08 46.10 27.90 7.50 Table 10. RMS Values of 25 Check Points for IKONOS
Panchromatic Image Using Point-based Technique with
Various Number of Control Points
Next, comparison between the developed program for point-
based projective equations and that of I/RASC was performed.
Results of the same 25 check points are summarized in Table
11. It confirms the accuracy of the developed program.
Then, the developed programs for line-based and combined
point/line-based projective equations were used with only 5
control lines and 3 control lines & 2 control points,
respectively. RMS values of the same 25 check points are also
presented in Table 11. It is clear that results of the line-based
and combined point/line-based techniques are equivalent to
those of point-based; however, with much fewer control
features. This proves the potential of using such line techniques
for the rectification of high-resolution imagery, and, thus,
188
International
reducing trem
needed.
RMS
X (m)
Y (m)
Table 11.
Panchron
4. CON!
]- Linear ima
to facilitat
abundant it
to automati
2- Projective
features,
combinatio
3- Least squ:
point, line
equations.
4- The devel
performing
satellite in
panchroma
mutlispectr
IKONOS p
5- Results of
were com
commercia
and Intergr
proved eq
developed |
6- Experime
point/line-t
equivalent
However, «
and 3 cont
point/line-l
results, wh
to achieve
This is ver
line-based
number of
still yield |
reduction i
cost and dt
7- Experime
features fo
imagery. $
line feature
Research is c
1-Use of ;
features. '
support
rectificatio
