The GPS is capable of providing very accurate position and 
velocity under ideal conditions. However, such conditions do not 
often exist. Independent GPS navigation requires at least four 
satellites. The major drawback GPS is the accuracy degradation 
due to poor satellite geometry, cycle slips, satellite outages, and 
dynamic lag during maneuvers. 
The INS measures linear acceleration and angular rates very 
accurately and with minimum time delay. For short time 
intervals, the integration of acceleration and angular rate results 
in extremely accurate velocity, position, and attitude with almost 
no noise or time lags. However, because the INS outputs are 
obtained by integration, they drift at low frequencies. To obtain 
very accurate outputs at all frequencies, the INS should be 
updated periodically using external measurements 
The function of each component of the VISAT system can be 
divided into primary and secondary function. In terms of primary 
functions, the camera cluster provides three-dimensional 
positioning with respect to the VISAT reference which in most 
cases is the perspective center of one of the cameras. The 
position of this reference with respect to the existing control is 
determined by differential GPS, while the camera orientation in 
three-dimensional space is given by the INS. The ABS system 
will trigger the cameras at constant distance using the VISAT 
controller trigger channel. In terms of secondary functions, the 
camera cluster provides redundancy (i.e., more than two images 
of the same object), the GPS controls the INS error propagation, 
and the INS, when used in positioning mode, bridges GPS 
outages, corrects GPS cycle slips, and gives precise interpolation 
between GPS fixes. The ABS data can be used to update the INS 
data if the GPS signal is blocked for periods more than the INS 
bridging level required to fix the GPS ambiguities ( half a cycle). 
The prerequisite for precise 3-D positioning from the VISAT 
video images is the systems calibration. System calibration 
includes the determination of the cameras inner orientation 
parameters, the relative location and orientation of the cameras, 
the relative location and orientation between the cameras and the 
navigation sensors. Surveying by VISAT, therefore, consists 
essentially of three parts: system calibration, position and 
orientation of the moving VISAT reference using GPS and INS 
data, positioning of objects in the road corridor with respect to 
the VISAT reference using two or more georeferenced camera 
images. 
2. GEOREFERENCING OFVIDEO IMAGES 
Georeferencing video images can be defined as the problem of 
transforming the 3-D coordinate vector r^ of the camera frame 
(c-frame) to the 3-D coordinate vector r'™ of the mapping frame 
(m-frame) in which the results are required. The m-frame can be 
any earth-fixed coordinate system such as curvilinear geodetic 
coordinates (latitude, longitude, height), UTM or 3TM 
coordinates. The georeferencing process can be described by the 
following formula (El-Sheimy, 1995) : 
m m ;" m b-c b 
r 7 ris (0 * S'* R, (Q(R, *r +a) (1) 
96 
Figure 2 shows the elements of the georeferencing formula, 
where : 
m is the coordinate vector of point (1) in the mapping 
frame (m-frame), 
m. (t) is the vector of interpolated coordinates of the INS in 
the m-frame, 
(t) is the time of exposure, i.e. the time of capturing the 
images. 
Ry (© is the interpolated rotation matrix between the INS 
body frame (b-frame) and the m-frame, as measured 
by the INS gyros. 
R, is the differential rotation between the camera 
coordinate frame (c-frame) and the INS b-frame 
r is the image coordinate of the object in the c-frame 
"mh is the offset between the INS center and the cameras 
si is a scale factor specific to a one point/one 
camera combination 
c - frame 
INS b-frame 
  
Figure 2 : The Georeferencing Concept 
The INS position me is the position resulting from the 
INS/GPS integration. In the event of a continuous loss of the 
GPS signal, the INS will be used in stand alone-mode to extend 
the mission. The high data rate of the INS, 50 Hz, facilitates the 
b 
interpolation of the camera coordinates. R c transforms the vector 
C m 
r from the c-frame to the b-frame. Ry (t) transforms the vector 
b b 
(n ° r+ 2 from the b-frame to the m-frame. R anda are 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B2. Vienna 1996 
detern 
Table 
expec 
proce: 
(1994 
The | 
coord 
image 
unknc 
soluti 
with ( 
  
  
  
High 
is an 
Syste 
parar 
They 
locati 
navig 
orien 
be us 
the 3 
The 
(GCI 
diam 
impk 
and ( 
and 
Coorc 
using 
been 
of il 
paraı