Full text: Proceedings International Workshop on Mobile Mapping Technology

3.1 Expert Knowledge in the Calibration Module 
The calibration module makes use of the calibration 
parameters stored in the expert knowledge database to 
estimate better parameters sing new calibration data. The 
calibration requires a number of Ground-Control-Points 
(GCP). A test-field of 70 circular reflective targets of 5- 
inch diameter were used. The process of identifying the 
target number and estimating its coordinate in the image 
space is a very time consuming. To overcome that, an 
expert module was implemented to automatically identify 
the targets number and then automatically detect its center 
based on its circular characteristics. The expert module, 
make use of the GPS/INS data and the targets 3-D 
coordinates in the automatic recognition process. The 
whole process can be summarized in three steps, as 
follows: 
Estimating the exterior orientation parameters of each 
camera using GPS/INS data and the previous 
knowledge on the relative orientation parameters 
between the cameras and the INS, as follows: 
R'Ji) = R‘ h (i).R b m 
, - -* 
'INS 1 
= r, 
• r. 
(1) 
Where, 
R m (i) is the rotation matrix between the mapping frame 
(m-frame) and the camera coordinate frame (c-frame) 
R L b (i) is the rotation matrix between the INS body frame 
(b-frame) and the camera coordinate frame (c-frame) 
r™ is the coordinates of camera (ci) in the mapping 
frame 
r^ s is the coordinates of the INS in the mapping frame 
is the coordinates of camera (ci) in the b-frame 
2. Estimating the approximate image coordinates of each 
target making use of the camera exterior orientation 
parameters and the targets 3-D coordinates. This is a 
back-projection problem of the photogrammetric 
collinearity condition. It should be mentioned that only, 
those targets that fall within the field of view of the 
camera would be back-projected into the image space 
(Figure 3). This is simply implemented by using the 
camera azimuth and field view as a condition for back- 
projeGtion. 
3. Using the estimated 2-D image coordinates of the 
targets, the system detect, calculate and validate the 
precise coordinates of the targets making use of their 
circular characteristics, for more details see El-Sheimy 
(19%). 
While it usually take 3-5 hours to calibrate the system, with 
the help of the expert module this time is reduced to 15-20 
minutes only. 
Figure 3: The Camera Field of View and its Relation to 
which Targets will be Back-projected 
3.2 Expert Knowledge in the Planning and the Mission 
Survey Modules 
Expert knowledge in these modules is required to automate 
and optimize the data acquisition stage as much as 
possible. The planning component is used to define an 
optimal survey mission using the knowledge acquired from 
previous missions for similar environments and equipment. 
The survey route, whether a straight or curved line, is 
divided into small line segments, which essentially follow 
the road pattern of the georeferenced media. The 
coordinates of the endpoints of these line segments are 
obtained from the georeferenced media over which the 
segments are superimposed. With the coordinates of the 
end points, the azimuth of each segment can be calculated, 
which is then stored along with the coordinates of the end 
points in a project database. This information will be used 
during the survey to guide the driver and then after the 
survey to evaluate the progress of the project. In the 
Mission Survey Module, the driver is guided to the defined 
waypoints, based on azimuth and distance information 
contained in this file. The cameras configuration and the 
distance between exposures are automatically selected in 
real-time during the survey. If the user choose an 
alternative route because of dense traffic or road closed, 
the system recalculated a new optimized route to respect 
the change. Alert messages ensure that the data collected is 
sufficient to provide the mission-required accuracy. Vital 
information is displayed to the survey crew on-line via the 
navigation control unit (Figure 4). It consists of the camera 
configuration in use, the number of satellites tracked, the 
azimuth and distance to the next waypoint, quality control 
alerts, etc. The information displayed on the navigation 
control unit is supplemented by spoken messages in critical 
situations, such as alerts or a change of the survey route. 
1-2-4
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.