dered 
in an 
about 
Ith of 
s is a 
ale is 
ler in 
using 
[rates 
point 
make 
ically 
oló' s 
^ the 
re of 
e two 
to the 
rome 
ically 
kable 
on in 
sarest 
patial 
nakes 
al or 
  
anner 
ect; 
we have made the angular stepwidth vary from 6 to 15 Mgon, 
optimising it according to the different parts’ features. A highest 
resolution (6 Mgon) has been applied to scan the most 
elaborated parts in order to obtain an accuracy of 1.0 cm 
5 
whereas a medium resolution (10-15 Mgon) with a 2.5 cm step 
has been considered enough to measure simple surfaces that did 
not present any ornaments. 
  
Fig. 7 — Lower part of altar's point cloud. 
The aim of our resarches was to obtain a full 3D model 
comprehending the texture information too. So that we 
simultaneously acquired both the geometrical data (by laser 
scanner) and the texture ones (by a digital camera). 
To reconstruct the whole model we used the 3D modelling 
software Surveyor developed by 3Dveritas. It registers point 
clouds in a global coordinate frame, integrates multiple scans, 
and generates high-precision polygonal models from the range 
data. 
  
Fig. 8 — Wireframe mesh 
In particular, the point cloud was processed in order to decimate 
some redundant informations (e.g. all the point lying in a plane) 
and to characterize all the mouldings of the columns’ basement, 
capitals, architrave, tympanum and statues. This way we 
obtained a very accurate triangulated wireframe mesh model 
ready to be mapped with digital images. 
Furthermore, choosing different crossing planes, we obtained 
various profiles of longitudinal, oblique and transversal cross 
sections. 
  
Fig. 9 — Textured mesh of the altar 
The 3D model obtained is a documentation as complete as 
possibile of the studied architectural object, since it can contain 
different kinds of information, both geometrical and about 
colour, x-ray, thermal, infrared. 
  
Fig. 10 — Textured mesh 
Nowadays the research is still in progress. During the next 
months, we will be able to obtain more complete results: in this 
framework we offer the processings handled so far. 
The gained experiences confirm that technology’s development 
and its application permit to obtain images metrically correct. 
So we can consider surveying operation as a repeatable 
scientific experiment, whose result can be compared to the 
previous one so as to be reliable: that is why scanner laser 
surveying can gain quality survey certification. 
—293—