have been superimposed four different corner col- 
umns and fifty-two small twin columns some of 
which have capitals decorated with grotesque and 
anthropomorphic figures. 
1- THE AIM OF THEWORK 
The research program was jointly initiated by the 
Istituto di Topografia of the University in Bologna 
and the Photogrammetric Laboratory of the Centro 
Cartografico of the School of Architecture in Venice. 
The purpose of this program is to compare non-stere- 
oscopical photogrammetric systems in view of a 
standardization which will regulate techniques and 
procedures of architectural survey. The ultimate goal 
is to clarify the limits of compatibility of the various 
systems. 
With this goal in mind, the cloister of Santo Stefano 
has been identified as an ideal field of experiment for 
the following reasons: 
- the architectural configuration on two levels sim- 
plifies the organization and the operations as the area 
on the first floor is protected and easily accessible; 
- the stone columns have bases and simply decorated 
capitals and do not present particular difficulties in 
the phase of restitution; : 
- the wall structure is visibly off axis and out of plumb. 
The brick surface has been restored in different 
periods with the insertion of stone elements and 
small brick decorations at the top of the walls. It will 
be therefore possible to verify all the systems at the 
various scales of representation. 
We foresee two phases of experimentation: 
- the first is to determine the accuracy of the various 
systems; 
- the second is to verify the suitability of the systems 
in the survey and representation of different architec- 
tural elements. 
In the first phase the intention is to operate only on 
marked targets to check the optical characteristics and 
the quality of the hardware and software, excluding 
  
  
  
  
  
  
  
  
  
L4 PITT Imm COE" 
Mag 
2 DEN 
Là NE 
Y 
X 
Fig. 3 
The topographic stations and the targets of the west and north 
facade 
for the time being the identification of homologous 
architectural points. The experimentation will be con- 
ducted on one hand rigorously respecting the appli- 
cable principles dictaded by the various systems, and 
on the other hand by imposing our own criteria to 
have a more homogeneous and therefore more com- 
parable results in the field of precision. 
We have therefore created a univocal reference sys- 
tem that is determined by the position of points sur- 
veyed with multiple forward resection and distance 
measurement. The points have been marked on half 
the perimeter of the cloister (fig. 1). This spacial dispo- 
sition allows one to evaluate the various photogram- 
metric systems with special attention to the possibility 
of detecting systematic effects produced by depth var- 
iations on the surveyed object. 
2- THE TEST OBJECT 
The area to survey is made up by the entire north fa- 
cade (fig. 2) and half of each of the adjacent facades. 
The surface has been marked with 245 targets (fig. 3); 
each target consists of a disk three centimeters in di- 
ameter made of a special reflective film upon which is 
printed a grid. The use of this reflective film allows to 
take the distance measure with EDM. 
The diameter of the disc and the grid correspond to 
the demands of precision in topographic measure- 
ment and in collimation during photogrammetric res- 
titution. 
The targets were distributed across the three facades 
according to a grid of about 1 x 0,7 m which was kept 
as regular as possible excepting of course in those areas 
impossible to mark due to the geometry of the archi- 
tectural elements. On the north facade 121 targets were 
distributed in 9 lines and 15 columns. 
The survey was then taken from four marked stations 
that establish the vertices of a plano-altimetric net- 
work with a trapezoid shape (fig. 3). Each vertex was 
marked following the principles of forced centering. 
The measures were taken with a total station Wild TC 
2000. The targets of the main facade were surveyed 
with linear and angular measurement from at least 
three stations while the adjacent facades were sur- 
veyed from the two stations on the opposite side. 
The results of the tridimensional adjustment in re- 
ference to the O-X-Y-Z system indicate that the stand- 
ard deviation of the coordinates Sx, Sy and Sz for 
nearly all the targets never exceeded 2 mm. A digital 
model of the north facade has been performed using 
the measured points pertaining the brick surface (fig. 
4). A series of color photographs were taken also with 
Galileo Veroplast metric camera and to complete the 
test, an analytical restitution has been made with 
Galileo Digicart 40 (fig. 5 - 6 - 7). 
3- THE USE OF THE TEST OBJECT 
The work will be articulated in two phases: 
1) evaluation of accuracy; 
2) evaluation of different survey procedures.