Full text: Proceedings, XXth congress (Part 5)

  
    
     
   
   
   
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
    
  
  
   
  
    
   
  
    
   
   
    
    
    
    
   
    
  
  
  
   
    
    
  
    
  
    
     
   
  
  
  
  
    
      
  
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5. Istanbul 2004 
  
with the photographic survey, a texture mapping of Baptistery 
to compare with the tri-dimensional model carried out 
photogrammetrically. 
  
Figure 1 — North side of the Baptistery with the portal. The 
Cathedral is in the left side. 
2. DESIGN AND SURVEY 
The project of the photogrammetric survey has been carried out 
trying to achieve an accuracy of centimetre on the control 
points. For this reason it is important to remember that the 
accuracy achieved by monoscopic multi-images restitution can 
be describe by the following formula (Mason. 1994): 
MIO 
e JE 
where: 
o. 1s the restitution accuracy 
mf is the photogrammetric scale (relation d/c) 
o; 1s the accuracy of measure on the image 
q is a geometry coefficient which can vary between 0.4 (very 
convergent shots) and 0.7 
k is the number of shots which are used to plot the single point. 
Oo 
[1] 
In relation with the classic bond valid for normal stereo-couples 
at the object (where the accuracy in the perpendicular direction 
to the shot is directly proportional to the distance square and to 
the measure accuracy of the plates, and in inverse relation to 
the base product of shot and principal distance), in the formula 
(1) is possible to consider both the number of images which cut 
on the object and the best geometry that is guaranteed by the 
convergent shots. ^ 
The camera used is a Nikon D100 with lens with principal 
distance c of 18mm and f/2.8. 
The digital back of 6 Mpixel has dimensions 23.4*15.6 mm”, 
which needs a dimension of pixel of 7.8 um and a si 
computed on the diagonal, of about 76?. 
The camera, of a no metric type, has been previously reamed by 
self calibration ona tri-dimensional test field. 
In figure 2 is visible the scheme of the 22 taken. It is remarkable 
that the followed logic scheme is that of surveying each of the 
eight sides of the Baptistery with a series of 3 images: a frontal 
and two lateral ones. The frontal one (position 3 in the figure 2), 
practically perpendicularly, is placed at the distance of about 
ght corner 
o 
16m, the two sides (10 and 11) are perpendicularly to the edge 
but convergent to the sides: the wider distance of survey allows 
a wider field of vision which allows to observe contemporary 2 
sides. : 
Palazzo del Comune 
Loggia dei Militi 
  
Figure 2 — a) Geometric design of the photogrammetric taken; 
b) position of the Ground Control Points. 
This scheme has been proposed for six out of the eight sides of 
Baptistery. The remaining two, leant to the structure of the 
Cathedral, have been surveyed by a series of convergent images 
which lie outside the scheme shown above. In practice, only the 
side on the right of the protirus has been put in the solid model, 
of the last side, distant to the base no more than 1.5 m from the 
Cathedral, and it has been possible to insert only the top side. 
Finally, due to the lack of dominant points of sight (Torrazzo 
was closed the day devoted to the survey) it was not possible to 
make a photogrammetric survey of the roof. 
To measure the positioning of the supporting points and of a 
certain number of control points (GCP) (figure 2b), it has been 
made a closed polygonal of three sides, from the top of which 
the 3D coordinates of 37 points were determined. 
For this aim it was used a total station Leica TC307 that for 
measures up to 80 m doesn't need a back-reflecting prism; this 
has allowed to accelerate the survey of the supporting points, 
chosen nearly entirely in correspondence with edges of the wall. 
Only three points used for the georeferentiation of laser shots 
are artificial targets. Plywood panels have been used to reach 
this aim and on these, on dark setting (at low reflectivity) IS 
drawn a circular white surface (with elevated reflectivity). The 
different radiometric answer between the light and the dark side, 
   
Internat 
TUE 
makes | 
from th 
the poir 
some m 
allows t 
  
Figu. 
The coo 
determi 
cm in y. 
photogr 
The sur 
terrestri: 
Milan. 
conditio 
40° whi 
2000 pc 
object i 
the flat 
For the 
of whicl 
photogr: 
loggia o 
taken af 
In photo 
the exte 
fixing o 
points, \ 
the pho 
orientati 
characte 
operator 
chromat 
the oth 
photogr: 
triangul: 
much le 
exists so 
field of 
approxir 
Zeng at 
homolog 
points ai 
In our | 
(www.p 
allow t 
orientati 
points, t 
(even if, 
It has be 
of that c 
allowed 
images : 
System |
	        
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.