1 0
T H E B U I L D I N G C O N S E R VAT I O N D I R E C T O R Y 2 0 1 6
T W E N T Y T H I R D E D I T I O N
1
PROFESS IONAL SERV I CES
Clients are struggling to specify clearly
their 3D data needs and the supplier/client
relationship is faltering. Sometimes the
client strays into the supplier’s territory by
specifying the technology (‘I want a laser
scan’) rather than defining the product (‘I need
a 3D wire-frame model to this specification’).
And as an industry we are sometimes guilty
of promoting the technology rather than the
solution.
However, clients in the heritage sector
are almost unanimously enthusiastic about
the potential of 3D information, particularly
because of its ability to communicate complex
information to the public. Heritage sector
clients can be demanding, keen to have
a comprehensive record of their site and
wanting 3D digital models for visualisation,
digital reconstruction and augmented reality.
Budgets, however, are usually very limited and
it is the task of the specialist survey supplier
to attempt to meet their aspirations at a
cost that is affordable and, in some cases, to
provide early advice on the practicalities and
implications of various methodologies.
Often, for the comprehensive
documentation of a heritage site, other field
datasets are obtained and this can include
archaeological excavation information in
addition to geophysical, ecological and
condition surveys. Historical evidence may be
available in a wide variety of forms including
maps, plans, documents, reports, photographs,
etchings and paintings, and these are valuable
where digital reconstruction is wanted – for
example, to display a particular period in a
site’s history (Figures 1 and 2).
3D DATA CAPTURE TECHNOLOGIES
Today specialist surveyors can use a number
of measurement technologies to create survey
imagery and 3D datasets to meet clients’
needs. Each has particular applications and
each has its advantages and disadvantages (in
addition to cost implications).
Laser scanning
Laser scanning captures a large amount
of 3D data, generating the now familiar
‘point cloud’ where each point has a distinct
Cartesian location, light intensity and colour.
Scanning can be done in poor light and even
at night. Data can be captured rapidly but the
instruments are relatively expensive. Another
disadvantage is that laser scanners can have
difficulty in obtaining data at height. Most
laser scanners require a stable platform during
scanning and using hoists and scissor lifts
can prove troublesome. For these reasons it
rarely provides a complete point cloud over a
complex heritage site and even with careful
planning, data voids can be significant.
In the heritage sector the point cloud is
rarely the final deliverable and laser scanning
may be an expensive solution depending on
the client’s actual needs.
Photogrammetry and Structure
fromMotion
Although it has a long history,
photogrammetry remains a powerful
means of obtaining point cloud data.
Stereo photographs are taken with precise
measurements and calibrated to allow
accurate orthographic imaging. More recent
Structure from Motion (SfM) techniques
follow the same principle but use a large
number of images from fixed-wing aircraft
or unmanned aerial vehicles (UAVs) and/
or taken from cameras on the ground,
on platforms, or mounted on poles that
allow comprehensive coverage of a site.
The images are processed to create a 3D
point cloud, mesh and textured mesh at a
resolution to suit the final output or use.
Instrument survey
Traditional land survey techniques using
total stations and GPS equipment are often
used to provide the control framework for the
laser scan and photogrammetric data capture
processes and to supplement the point cloud
information captured by other methods.
Only the points selected in the field to
create the final output are measured. Figure 3
shows the vectors captured by a reflectorless
total station to define a building facade.
Subsequent modelling and photographic
texture were added to the model to create the
end-product. For a client viewing the end-
product, it would be difficult to determine the
mix of methodologies used, but each provides
distinct types of data which can be extracted
and used for a variety of functions.
Traditional measured survey
In spite of the advanced survey technologies
available, there are many situations where a
very low-tech approach is equally effective
for the survey of historic buildings and may
be more appropriate and cost-effective. This
is particularly true for smaller, more complex
or irregular structures, for the recording of
individual architectural details or for sites
where survey needs to take place in stages
– before and then during a programme of
building works.
It is a distinct advantage that survey is
carried out in detail by those who are also
responsible for assessing and understanding
the building, so analysis can be made
and developed as the survey proceeds in
a continuous and seamless process. Thus
the ability to carry out accurate hand
measurement and good drafting skills
developed over many years are essential tools
for many buildings archaeologists.
Hand survey can involve a spectrum
of activities – sketched individual details,
measured sketches, hand-measurement of
features which are then drawn directly onto
metric paper, the marking-up of existing
drawings or imagery, and so on. And most
commissioned survey work on historic
buildings, whether digital or by hand, still
involves a significant component of close
secondary scrutiny on site by the archaeologist
during which survey imagery is subject to
correction and refinement, and the layering-
on of detailed analytical and interpretive
information. There is little substitute for direct
one-to-one contact between archaeological
surveyors and their subjects.
Today the outputs of such drawn surveys
are usually converted into digital imagery in
the office and can then be merged with most
other digital graphics media.
ANALYSIS AND UNDERSTANDING
None of these data capture techniques in
themselves generate the final output or
analytical record of a building. The imagery
requires processing, presentation to the
required standard and the addition of further
layers of interpretive information. High
quality survey data can obviously support and
speed up this process but it is only ever the
base for a full analytical visual inspection and
detailed record.
An archaeological analysis of a historic
building typically gathers a great deal of
additional qualitative information, usually
through a process of careful visual inspection
of the subject.
It is routine for this information to
be superimposed on the graphic imagery
as additional layers or linked information
(Figure 6). This includes constructional
information, material types, information
relating to phasing and function, catalogue
references for information on specific features,
mapping of mortar types, petrological
assessment and so on. There are also inputs
from other specialists such as moisture
mapping, rot surveys, the plotting of
structural failure and paint analysis data.
Figure 3 A 3d textured model created from total station survey measurements