43 
e descriptive 
illy used in 
.tion. If the link 
oe so dense to 
could lead to a 
up to a stage 
ain interest) its 
rfaces, in most 
ssing office, to 
al design of the 
nly known, 
le selection of 
3 system and 
re adequate to 
, the reference 
parallel to the 
jdinal section, 
and final spans 
representation 
as a rule, the 
buildings have 
undergone modifications that testimony, above all, different 
construction stages; resulting in the lengths related to the 
almost parallel planes being reduced while almost 
orthogonal planes would provoke shifting between different 
faces inserting them in foreshortening. It’s interesting to 
point out the way in which the projection system risks to 
include ambiguity in interpreting measurement result, even 
though it has been made in best possible way. 
The plenty of acquired data has brought up, during the 
processing, a series of thoughts on their possible use, on 
future developments of these methodologies and on the 
corrections to be applied during data acquisition process, 
in order to quicken up the operation of image creation. 
Thanks to the relative speed and simplicity of data 
acquisition and to ever common calculation programs, it 
is possible to transfer to the processing phases the 
simplified geometric assumptions, deriving from the 
number of acquired points: two points are needed to define 
a line, three to define an arc, four for a sphere, but is it 
really a line, a circumference arc, a spherical element? 
A search for mathematical relations that could 
demonstrate as closer as possible the real architectural 
shape out of aggregate of the measured points, minimizing 
the error, until recently it was more easily applicable to 
photogrammetric images and now it is possible also to 
start with the data acquired by topographical methods 
provided that they have been subordinated to a preliminary 
and precedent surveying project containing the detail 
points. 
There is no need to make an indistinct number of points 
but it is necessary to collect the date that reflect reasonable 
choice that could be rebated in the calculated stage in 
which the data could be composed to the geometric 
theories reflecting the concept itself of the architectural 
object. 
The result could be compared using general relations, 
cone, quadratic, ellipsoid, resolving equation system that 
a being a formal preliminary proposition. 
In this case a computer becomes indispensable for the 
comparison of the similar forms and for eventual reiteration 
and search for the most adequate ones. 
Studying the partition of errors and their relative value, it 
will be possible to observe directly the deformations of an 
architectural structure and to furnish necessary information 
on the conservation stage of the building. 
REFERENCES 
Porter A. K., Lombard Architecture, 3° voi., New Haven, 
London - Oxford. 1917. 
Bezoari G., Guzzetti F., Impianto generale di un rilievo 
architettonico. Recuperare, 2/92. 
Astori B., Bezoari G., Guzzetti F., Analogue and digitai 
methods in architectural photogrammetry. Atti 
dell’lnternational Congress of Photogrammetry and 
Mathematical model software are very useful instrument 
for managing such a bulk of information and maybe they 
are opening a new prospective that would allow us to cross 
over the limits of the conventional representation. 
5. IMAGE AND DESIGN 
To the necessary critical selection that makes the survey 
and the graphic representation an instrument of immense 
value, it is also to be added the valuable characteristics of 
a photo. A passage from analogical to digital allows 
nowadays more flexible use of a photo imaging that 
releases a “global memory” to analytical designs achieved 
by topographic measurement. 
Aiming at a reduction of a field work and in order to document 
in a best possible way the metric difference of the surfaces, 
digital rectification of the walls within the nave has been 
made thanks to which it was possible to define in detail 
the values achieved outlining the areas where the contours 
have been measured by topographic use of points (figg. 5, 
6, 7, 8). 
For each single span, it was prepared, together with the 
profile measurement, acquisition of at least five visible 
and accurately monographed control points aimed at 
creating the topographic support to digital rectification.The 
filming has been made with a Rollei 6008, with 50 mm 
zoom, supplied with a reticle with 9 crosspoints. Long 
periods of exposure, due to the lack of light, required a 
tripod. 
For the colour scanning of the photograms provided in that 
way, a resolution of 800 dpi has been used; the prospective 
correction of the imaging has been made using a 
commercial software. 
Thanks to the control points acquired in one single 
reference system, it was possible to set a mosaic of 
photograms although it lacked the overlapping areas 
between the adjacent images as they were interrupted by 
the veils and transversal arches (obviously not belonging 
to average rectification plan). 
Through the vectorialization of digital results the 
representation of elevations has been completed. 
The joined use of the two systems allows a minor 
geometrical abstraction reducing the gap between what 
has been measured with the measuring operations and 
what has been calculated during data processing (fig.9). 
Remote Sensing. Washington, 4-15 agosto 1992, 
Commissione V. 
Guzzetti F., Tucci G.,// nuovo rilievo topografico per il 
recupero. Recupero & Conservazione 14/97. 
Guzzetti F„ Rilievo tridimensionale di una torre finalizzato 
all'analisi strutturale agli elementi finiti. Rivista del 
Dipartimento del territorio, n. 2 1997, Roma.