International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
Figure 2. Overview of block of Palermo with all the ground 
point. 
3. DESCRIPTION OF THE EXPERIMENTATION 
The experimentation was carried out comparing the direct 
exterior orientation parameters with those calculated by means 
of an aerial triangulation (AT), and using some GCPs to check 
their accuracy in different configurations taken into 
consideration. 
At first the parameters directly provided by Applanix system 
(APP) were compared with those calculated by an aerial 
triangulation (AT) performed introducing the direct orientation 
parameters and measuring an adequate number of tie points 
(ATI). 
À second comparison was done by a traditional AT, using some 
GCPs without exterior orientation parameters (AT2). 
Further comparisons were done by AT operations carried out 
integrating the direct orientation parameters with some tie 
points and GCPs; particularly, two configurations were 
considered: the one with a single GCP located in the middle of 
the blocks (AT3a), another with four GCPs in the corner of 
blocks (AT3b). 
For the Catania flight other AT operations were also carried 
out, with only the perspective centres and four (AT4a) or seven 
  
  
  
  
i 
  
  
Figure 3. Overview of block of Catania with all the ground 
point. 
GCPs (AT4b). 
Various procedures of orientation were executed by Inpho 
MATCH-AT software and Bea Systems Socet Set 4.4 software. 
From the exterior orientation parameters provided by Applanix 
‘system and ones calculated in the different types of AT taken 
into examination, some digital orthophotos were produced. 
From the relative comparison, made in correspondence of well 
identifiable image points by means of a Arcview environment 
specially compiled software, it has been possible to draw 
conclusions about the influence of the parameters variations on 
the orthophotos accuracy. 
DEMs necessary for making orthophotos were derived from 
available maps: at scale of 1:10000, for the city of Palermo, and 
1:2000, for Catania. 
3.1 The Palermo project 
As regards AT2 operation conducted in Palermo, 9 tie points 
per photograph were observed, choosing points visible on 
several photographs, between adjacent models and strips. 63 tie 
points were observed above. To calculate exterior orientation 
parameters, only 12 of 47 points surveyed by GPS were 
employed as GCPs, and the remaining points as CPs. 
Table I shows the aggregate results of the comparison between 
the exterior orientation parameters provided by Applanix 
system (APP) and ones calculated by Socet Set software for 
different configurations taken into account. The comparison 
with AT2 operation highlights the most relevant discrepancies, 
even if not particularly significant; in the other cases the 
discrepancies are of a smaller order of magnitude and 
independent of the GCPs presence and number. 
  
  
Inter 
———— 
  
  
  
  
  
  
  
  
X Y Z Omega Phi Kappa 
Mean 0.000 0.000 0.000 -0.0003 0.0007 0.0024 
APP-AT1 
St. Dev. 0.002 0.001 0.003 0.0035 0.0041 0.0029 
Mean 0.067 0.045 0.130 -0.0041 0.0026 0.0040 
APP-AT2 
St. Dev. 0.589 1.114 0.206 0.0221 0.0129 0.0049 
Mean 0.000 0.000 0.000 -0.0003 0.0017 0.0022 
APP-AT3a 
St. Dev. 0.002 0.001 0.003 0.0030 0.0045 0.0030 
Mean 0.000 0.000 0.000 -0.0043 0.0024 0.0047 
APP-AT3b 
St. Dev. 0.001 0.001 0.003 0.0036 0.0033 0.0057 
  
  
  
  
  
  
  
  
  
  
Table I. Statistical parameters of the comparison from exterior orientation parameters for Palermo project (X, Y, 
Z, in meters — omega, phi e kappa in grad) 
576