Chapter 6

Terrestrial Laser Scanning

Since the early 2000s terrestrial laser scanning has evolved from a research and
development (R&D) topic to a geo-data technology, which is commercially offered
by a multitude of land surveying companies and other service providers all over
the world. The technology is primarily used for the rapid acquisition of three-
dimensional (3D) information of a variety of topographic and industrial objects.
Cultural heritage, bridges, plants, cars, coastal cliffs, highways and traffic colli-
sion damage, all can be accurately modelled and documented with laser technology.
Lidar is without doubt the most successful data-acquisition technique introduced in
the last decade. As an acronym of Light Detection and Ranging, some prefer to read
Lidar as Laser Imaging Detection and Ranging – the term has become a ‘proper
name’ – spelled like your own first and surname with the initial letter the only
capital.
Basically, all Lidar scanners measure range and intensity of terrain points hit by
the laser beam. The first task is to convert the raw data into positions – three coor-
dinates for each point – in a geodetic reference system. The resulting point-cloud is
the basis for further processing, including filtering, visualisation, classification and
analysis, or other manipulation.
Lidar instruments can be mounted on a tripod positioned over the ground
for capturing, at street view, the surface of objects in the surrounding, such as
bridges, dams, building facades, trees or cultural heritage sites (Fig. 6.1). In these
ground-based cases the technology is called terrestrial laser scanning (TLS). Lidar
instruments can also be placed in moving platforms such as aircrafts, helicopters,
cars or vessels. When placed in a flying platform the technology is usually referred
to as airborne Lidar or airborne laser scanning (Lemmens, 2007a). When the laser
scanner is placed in a car, van or boat, the technology is called mobile laser scanning,
terrestrial mobile mapping or, more often, mobile mapping. In recent years mobile
mapping has become a rapidly emerging technology, particularly for accurately
mapping roads and highways. The biggest advantage of mobile mapping is avoiding
closure of the road during the time surveyors would manually measure the trajec-
tory and the extreme rapid acquisition of millions of points. This chapter presents
the principles of terrestrial laser scanning and provides an overview of application
areas. Mobile mapping systems are also discussed. Airborne Lidar is treated in a

M. Lemmens, Geo-information, Geotechnologies and the Environment 5, 101

DOI 10.1007/978-94-007-1667-4_6, 
C Springer Science+Business Media B.V. 2011
102 6 Terrestrial Laser Scanning

Fig. 6.1 A Leica Geosystem laser scanner at work for heritage documentation in Africa

separate chapter (Chapter 8) after discussion of the basics of photogrammetry, as

airborne Lidar and aerial photogrammetry show many similarities especially with
respect to geo-referencing and applications. In-depth treatment of theory and prin-
ciples of Lidar technology can be found in three specialised textbooks, which have
been recently published and written by a wide range of scientists and other experts:
Shan and Toth (2008), Heritage and Large (2009), and Vosselman and Maas (2010).
The technical specifications and development of commercial terrestrial laser sensors
are listed on a regular basis in Lemmens (2004, 2007d, 2009, 2010).
Three-dimensional (3D) modelling aims at capturing all geometrical detail of
objects, both the exterior parts and the interior ones, and representing these features
with high-resolution triangular meshes for accurate documentation and photo-
realistic visualisation (El-Hakim et al., 2007, 2008). Terrestrial Laser Scanning
makes possible the swift measurement of millions of points by automatically scan-
ning the scene at high speed. In the resulting dense point-cloud objects can be easily
identified, allowing the creation of 3D-models of a wide range of objects with a level
of detail impossible to achieve within a reasonable space of time using traditional
technologies. Practical tests have shown that terrestrial laser scanning compared to
conventional land surveying methods, can be up to 15 times faster for creating dig-
ital elevation models of open pit mines (Conforti, 2010). The high level of detail
with surveying accuracy gives users access to a wealth of information. The prod-
ucts based on dense point-clouds are used by a multitude of professionals such as
cultural archaeologists (Fig. 6.1), architects, facility managers, chemical engineers,
construction engineers and plant engineers.
6.1 Basics of Laser 103

6.1 Basics of Laser

A laser is a device that creates a narrow, intense beam of coherent and monochro-
matic light generated by particles (atoms or molecules) that emit radiation (Siegman,
1986; Wilson and Hawkes, 1987; Koechner, 1992; Silfvast, 1996) The easiest way
to understand the working of a laser device is by considering its two-level system:
particles have only two energy levels, an upper and a lower state, separated by some
energy difference. Electrons in the lower state can be excited to the upper state,
usually by heat. Upon returning to a lower state the particle emits light. Usually
particles behave independently of each other and the resulting light can be of any
wavelength (colour). However, the emission of radiation can be stimulated to pro-
duce light of a particular wavelength. This occurs when a particle in the upper state
interacts with a photon matching the energy separation of the levels; the particle may
decay, emitting another photon with the same phase and wavelength as the incident
photon, resulting in two photons emitted. This process is known as stimulated emis-
sion. The resulting beam consists of light of a single wavelength; the light is said
to be monochromatic. Wavelength is determined by the amount of energy released
when the electron falls to a lower state. The beam is also coherent, meaning that the
photons move in step with each other; they have the same phase. The divergence
of the beam is small, resulting in a very concentrated and narrow ray of electro-
magnetic energy. A flashlight, on the other hand, releases weak and diffuse light in
many directions. Lasers differ greatly in properties, as well as in wavelength, size
and efficiency.
Albert Einstein recognised the existence of stimulated emission in 1917, but not
until the 1950s were ways found to transfer the theory into practice (Bertolotti,
1983). In 1954 Charles Townes, researcher at Bell Labs US and, independently,
Basov and Prokorov in Russia, suggested a practical method of achieving lasing
using ammonia gas, which produced amplified microwave radiation instead of vis-
ible light (called a Maser). For this the scientists shared the 1964 Nobel Prize
for Physics. In 1958 Townes and Arthur Schawlow in a paper entitled Infrared
and Optical Masers calculated the conditions needed to produce visible laser light.
T. Maiman used a ruby crystal to demonstrate the first true laser in 1960.
Compared to other light sources, the light produced by lasers is in general highly
monochromatic, directional, powerful and coherent. A price to be paid for these
beneficial properties is that lasers can harm living tissue (Lemmens, 2010). As laser
scanners often operate in the presence of people, such as streets in cities and plants,
safety is an important issue. Since the early 1970s lasers have been subject to safety
classification based on their potential for causing damage to eye and skin. The ini-
tial classification system of six classes was slightly revised and extended to include
seven classes in 2002. In the old system the US indicated classes by Roman numer-
als, sometimes appended with letters (I, II, IIA, IIIA, IIIB and IV), whereas the
EU used Arabic numerals (1–4), also appended by letter; the latter system is used
worldwide in the revised version. The higher the numeral, the more injury the laser
can cause: 1, 1M, 2, 2M, 3R, 3B and 4. Most TLS systems are assigned to safety
104 6 Terrestrial Laser Scanning

class 3R; in the old system this would be IIIA, meaning ‘safe if handled carefully’.
Safety depends on maximum output power, laser principle and wavelength. The out-
put power of phase-shift scanners operating in the visible wavelengths should not
exceed 5 mW. Other maxima apply for other wavelengths and for pulsed lasers.

6.2 Measurement Principles

Laser scanners are active sensors that emit laser beams for measuring the distances
to objects without human/object contact; ‘active’ means that the sensors themselves
emit electromagnetic (EM) energy (Fig. 6.2). Figure 6.3 shows a typical output of a
street scene recorded by a laser scanner and derived CAD drawing.

Fig. 6.2 Principle of

Terrestrial Laser Scanning;
from measured range and
scan angle combined with
position and orientation of the
scanner the three-dimensional
coordinates of the points hit
by laser beams can be
computed, resulting in a point
cloud of 3D coordinates

Fig. 6.3 Part of a street in the historical area of Leicester, UK, captured by a laser scanner mounted
on a car; point cloud computed from the raw data (left) and derived CAD drawing
6.2 Measurement Principles 105

6.2.1 Time-of-Flight Versus Phase-Shift

Measuring with laser sensors may be based on four principles (Colombo and
Marana, 2010):

– Phase shift: waves are modulated in width or frequency; width modulation is sen-
sitive to sharp discontinuities in shape or reflectance of the object, while frequency
modulation provides reliable measurements even when return energy is low.
– Pulse measurements, also coined time-of-flight: pulses are emitted and their travel
time to and back from the object is measured.
– Optical triangulation, for short-range applications and small objects.
– Interferometry, which offers very high precision and is used in indoor industrial
metrology.

The first two laser principles, phase shift and pulse measurements, are commonly
used in TLS systems for outdoor applications (Lemmens, 2004). In phase-shift
technology the sensor continuously emits beams which are modulated as sine waves
(Figs. 6.4 and 6.5). The phase of the reflected part is measured and compared to the
phase of the outgoing one, and distance then calculated from the difference in phase
(phase-shift). In the time-of-flight technology a pulse is emitted in the direction of
the object; the time taken by the part of the pulse reflected back to reach the instru-
ment is measured. Distance is calculated by multiplying this travel time by the speed
of light and dividing the result by two.

Fig. 6.4 Two terrestrial laser scanners available on the market based on the range measurement
principle of phase-shift, Z&F Imager 2006 (left) and FARO LS880
106 6 Terrestrial Laser Scanning

Fig. 6.5 Two terrestrial lasers based on the principle of time-of-flight range measurement: Optech
ILRIS (left) and Riegl LMS Z420i

6.2.2 Range

One of the most important features of a TLS instrument is measurement range

because range determines to a large extent types of application (Lemmens, 2004).
A distinction can be made between short-range (up to 25 m), medium-range (up to
250 m) and long-range (larger than 250 m). Phase-based scanners, on the market
since the early 1990s, were initially aimed at close-range, high-accuracy industrial
applications. They are characterised by a precision ranging from sub-millimetre to
sub-centimetre level, and high scan rates of up to half a million points per second
resulting in high density. However, these favourable numbers come at the cost of
limits to range, which is less than 100 m. In contrast, time-of-flight systems may
measure distances up to 1 km and even more, but their precision usually ranges
from sub-centimetre to centimetre level, while scan rates are at the 10,000 points
per second level.
Another important feature of a TLS instrument is spatial resolution; this param-
eter determines the level of detail that can be obtained from a laser point cloud
and depends on the range (distance to the object), sampling interval and width of
the laser beam (Lichti and Jamtsho, 2006). Usually the sampling interval is taken
as indicator of spatial resolution. However, a high sampling interval does not nec-
essarily mean high spatial resolution; if the beam width is rather large, a high
sampling interval will cause that the footprints of the laser beams as projected on
object surfaces do overlap, reducing the effective spatial resolution which one would
expect by considering sampling interval alone. Lichti and Jamtsho (2006) propose a
new, more appropriate measure: the effective instantaneous field of view (EIFOV).

6.2.3 TLS as a Surveying Instrument

Many TLS systems appear less than perfectly suited to daily survey applications
because the equipment is big, difficult to move around and requires an external
6.2 Measurement Principles 107

power supply and computer support for operation (Neubauer, 2007). In many
applications, these characteristics may fade into insignificance compared with the
benefits; but survey workflow demands versatile set-ups. Therefore manufacturers
of surveying equipment such as Topcon and Trimble introduced TLS instruments
that can be carried around, set up and operated as a total-station. The battery is
internal and the camera is also built in, while the control panel is part of the system.
Most TLS systems do have in built imaging functionality, but some manufacturers
choose a non-integrated-camera laser scanner, offering instead a camera mounted
on top of the instrument. A scanner without integrated camera improves user flex-
ibility. For example, users may want to apply a certain camera type because they
are used to it or because their applications require a specialised camera, perhaps a
calibrated metric one. It may also be desirable to change lenses depending on type
of job (Fig. 6.6).

Fig. 6.6 The Riegl VZ-400 with a Nikon digital camera mounted on top (left) emits 122,000
near infrared pulses per seconds, weights 12 kg and measures ranges up to 600 m. Right: Leica
ScanStation C10 emits 50,000 green laser pulses per second, weights 14 kg and measures ranges up
300 m green. Both instruments were demonstrated at the European Lidar Mapping Forum (ELMF),
December 2010, The Hague, the Netherlands
108 6 Terrestrial Laser Scanning

6.3 Processing Software

The successful use of laser-scanner systems depends not only on the characteris-
tics of the data-collection instruments themselves but also on the capabilities of the
processing software necessary to obtain meaningful information after field acquisi-
tion of the 3D point cloud. The diversity of characteristics shown by the different
software packages, available on the commercial market, demonstrates that 3D-laser
scanning is a promising surveying technology, which is in the process of ongoing
and rapid development. Some of the software packages are developed for specialised
applications, while others offer a full suite of point-cloud processing tools from
scan registration and georeferencing through to feature extraction (Lemmens, 2006).
They are specifically developed to support manufacture – specific scanners. There
are also packages that offer full support for the entire laser-scanning process, right
through to reverse engineering and data comparison, and are not associated with a
particular type of scanner. 3D-laser scanning was not initially specifically developed
as a technology to support the needs of the land surveyor. However, manufacturers
of surveying instruments have recognised the attention being paid by surveyors to
the technology. In response, these manufacturers are increasingly adapting the tech-
nology, both instruments and processing software, to the needs of the surveyor. As a
result, new features are steadily added in support of improving surveying workflows.

6.4 Comparison with Total Stations

Laser scanners are often compared to reflectorless (robotic) total stations and, as far
as the measurement principle is concerned, this is fine. Both instruments measure
distances using pulsed laser light or phase shifts. This section treats resemblances
and differences between TLS systems and total stations, equipment which – together
with GNSS receivers – belong to the standard equipment of the land surveyor based
on the analysis carried out in Lemmens (2007c).

6.4.1 Measurement Principles and Precision

Total stations can achieve higher precision because many measurements, even up
to thousands, to the same point are taken and averaged, and thus improving accu-
racy, while laser scanners measure each distance only a few times, sometimes just
once. To determine 3D coordinates of object points, the coordinates of the position
of the instrument have to be known as well as the horizontal and vertical angles
of each outgoing beam. The 3D coordinates of each point are calculated in a local
or national reference system, from the laser distance, the known X, Y, Z coordi-
nates of the instrument, and horizontal and vertical angles of each outgoing laser
beam. The coordinates of the position of total stations are usually determined by
centring the instrument above a known point. The position of a laser scanner is
usually determined indirectly by placing special targets the three coordinates of
6.4 Comparison with Total Stations 109

which are measured using traditional survey instruments such as total stations. The
similarity in measurement technology inspired manufacturers of traditional survey
instruments to modify their total stations into a quasi laser scanner able automat-
ically to scan areas of interest at predefined intervals, despite scanning rate being
significantly lower than that of laser scanners.

6.4.2 Blind Sampling

At product level, similarity bounces. A land surveyor is used to interpreting a scene
as a collection of characteristic points each of which has to be measured individ-
ually; connecting the characteristic points correctly allows reconstruction of the
boundary of the object. Using a laser scanner, no selection of individual points takes
place during scanning; it is a matter of chance which points are hit by the laser beam.
As a result, unwanted objects, such as crossing pedestrians, are also captured. The
actual measuring is done in the office by fitting geometric primitives such as lines,
planes, cylinders and spheres through parts of the point-cloud. Boundaries and char-
acteristic points are computed from intersecting neighbouring geometric primitives.
Laser scanners can collect points for hours on end without human intervention, mak-
ing them particularly suitable for capturing complex objects and operating in hostile
environments such as nuclear plants, where placing the instrument should be done
pretty smartly.

6.4.3 Time-Efficiency

Laser scanners and total stations are also compared at time-efficiency level. Indeed,
surveying with a total station is only feasible when the object can be modelled by
a limited number of characteristic points. Laser scanning enables capturing scenes
consisting of objects of complex shape, such as chemical plants, cultural-heritage
and traffic-accident sites. Some vendors of laser scanners posit the idea that if a
laser scanner acquires 8,000 points per second while it takes a survey team 10 s
to measure a single point, using the scanner is equivalent to working with not one
but 80,000 teams; a faulty and misleading comparison. Surveyors select their points
intelligibly, while laser scanners take points blindly, without identification, interpre-
tation and selection. These activities have later to be carried out in the office. As a
rule of thumb one may state that the processing of industrial objects, such as plants,
will take approximately the same time as the time spend to capturing the data in the
field (1:1 field to processing ratio). But as far it concerns heritage documentation a
field-to-processing ration of 1:10 is more realistic, that is one day of data captur-
ing in the field requires up to ten man-days data-processing in the office (Rüther,
2007; Lemmens, 2007b). As such, laser scanning bears more comparison with pho-
togrammetry than with surveying. Laser scanning may thus bring land surveyors
and photogrammetrists closer together.
110 6 Terrestrial Laser Scanning

6.5 Comparison with Terrestrial Photogrammetry

When comparing the performance characteristics of TLS with terrestrial photogram-
metry techniques the pros and cons can be summarised as follows (Lemmens and
van den Heuvel, 2001):

1. On-site recording by TLS is independent of the presence of texture on the object.

Texture is particularly essential when digital surface models of the objects have
to be created from stereo images. The presence of texture in photos is however
not required when edges of objects are measured to create for example as-built
models.
2. Images can be taken faster and capture both geometry and texture, while TLS
systems capture geometry only.
3. By using two or more cameras that can be simultaneously exposed, photogram-
metry enables instantaneous recording of a scene. Consequently the recording
of dynamic objects is possible. Notwithstanding that the scanning speed of TLS
systems is very high, the scene is captured sequentially, point by point, and the
recording of a scene may take up to 20 min or even longer. So, in contrast to
photogrammetry, dynamic objects, such as human bodies, cannot be recorded by
TLS without distortions in the data.
4. Since TLS is an active system, generating its own light, while photogrammetry
makes use of light generated by external sources (sun, lamp), light conditions are
less critical for TLS.
5. Although manufacturers of surveying equipment are working on instruments
that can be carried around, set up and operated as a total-station, today’s TLS
systems appear less than perfectly suited to daily survey applications because
the equipment is big, difficult to move around and uses a lot of power neces-
sitating carrying around heavy batteries. However, their robustness makes them
suitable for use in harsh environments, such as deserts and tropical rain forest.
Recording scenes by photogrammetric techniques is more flexible because cam-
eras are more portable (less heavy) than TLS instruments, while one does not
necessarily have to mount them on tripods.
6. Although the prices of high-precision TLS systems are gradually decreasing
as a result of becoming a matured technology, they are more expensive that
photogrammetric systems with comparable precision.
7. The creation of digital surface models, necessary for visualisation and rendering
purposes, amongst other things, can be done directly and automatically from
the recorded TLS data. Photogrammetric images first require a matching step
in order to create a set of 3D point, for which specialised matching and editing
software is needed (see Section 7.5).

The above shows that the characteristics of TLS and terrestrial photogrammetry do
not compete with each other; they are complementary in nature. Since they supple-
ment each other, TLS and digital cameras are often used alongside when capturing
complex scenes.
6.6 Integration with Digital Cameras 111

6.6 Integration with Digital Cameras

Technology is moving towards the development of hybrid measurement system con-
sisting of a TLS device solidly integrated with a calibrated digital camera. Such new
devices collect 3D point-clouds of objects and, simultaneously, digital images of
the same object. They thus integrate surveying with photogrammetry and imaging.
These systems are useful both for measuring 3D building points and for directly ren-
dering 3D models with high-definition photo-texture and are useful for architectural
surveying, 3D modelling of monumental buildings and heritage documentation. The
combination of point cloud and image can be done highly automatically because
the position and orientation of scanner and camera are mutually known with high
accuracy avoiding manual identifying a set of homologous points between models
generated from the laser scans and each image. Colombo and Marana (2007), both
from the University of Bergamo, Italy, tested the information content and perfor-
mance of such hybrid systems by 3D modelling of S. Maria Maggiore, the most
important church in Bergamo and found good results (Fig. 6.7).

Fig. 6.7 3D Point models of the interiors of S. Maria Maggiore, Bergamo, Italy (top) and
orthoimages of decorated intrados of dome and vaults, with line plots in overlay (bottom)
112 6 Terrestrial Laser Scanning

Fig. 6.8 Venice palace (Italy) laser-scan data merged with bathymetric point-cloud data

They also conclude that the present manufactory standards for scanners have
to be improved as well as the processing software. Although scenes can be cap-
tured highly automatically, the in-office information extraction is tedious and
labour-intensive because automation level of the processing software is low.
TLS instruments are not only used in conjunction with cameras but also with
other types of sensors, such as sonar. The combination of 3D-laser scanning and
side-scan sonar can be very beneficial for mapping complicated waterside areas,
because the two systems are complementary as Byham et al. (2007) and Bacciocchi
et al. (2009) demonstrated (Fig. 6.8).

6.7 Scene Monitoring

Vendors of terrestrial Lidar scanners often proudly present the operation of their
instruments in the field as ‘a piece of cake’. And they are right; it is convenient. One
just needs to place the instrument in front of the object, level it roughly, enter the
area to be captured and push the button. It is as easy as that, at least as far as the
instrument side of the survey is concerned. But the snag lies not in operating the
instrument but in the prelude to this, and in scene monitoring during capture.

6.7.1 Intervening Objects

Before placing the instrument the operator has to ensure that no objects intervene
between instrument and scene that might occlude essential parts of it. This means
6.7 Scene Monitoring 113

no cars, lampposts, traffic signs, vegetation, people or donkeys (Figs. 6.9 and 6.10).
Potential sources of failure may also be hidden in the scene; lapses may result
especially from the reflectance characteristics of the surfaces of objects.
Hitting a surface, a laser beam may interact with it in three ways: it may be
reflected, absorbed or transmitted (Rees, 2001). Only reflected beams will reach
the instrument and thus be of use, but one reflection is better than another. Ideally,
a surface behaves as a diffuse reflector, so that the resulting reflections are of
like strength in all directions. In this case most of the signal returned from the
surface reaches the instrument. But when parts of an object have a specular sur-
face the reflection is deflected and little or no signal is returned to the instrument.
Therefore mirrors, shiny metal and brackets of neon lights present in the scene have
to be removed or covered. Signal strength, recorded in addition to time of flight of
the laser beam, provides the operator with a helpful means of detecting lapses in
reflectance during processing of point-clouds. The intensity of the reflections is an
important factor determining the accuracy of TLS point measurements. The inten-
sity mainly depends on three aspects: distance from the instrument to the object,
reflection properties of the surface of the object and angle of incidence (Lee et al.,
2010).
Laser beams may be transmitted through windows; this can happen with build-
ings, and results in the recording of objects on the other side of the glass. To avoid
problems in reconstructing the final scene, windows have to be covered, or special

Fig. 6.9 Trees, fences, and

also unmovable donkeys
(bottom left) may conceal the
scene from the laser scanner,
impeding capture
114 6 Terrestrial Laser Scanning

Fig. 6.10 Crossing of pedestrians and bikers, while the scene is being captured, causes appearance
of spikes and other funny things

caution must be exercised in processing the point cloud. Highway police, when
using scanners to record the site of an accident, spray powder on car windows to
prevent the scanner ‘looking’ through them. Pedestrians tend to cross a scene whilst
it is being captured, and the same is true of animals, cars and bikes. The surveyor
must thus close off the scene area to all traffic before pushing the button. This might
sound self-evident, but in practice closing the scene to moving objects often proves
much more problematic than it sounds.

6.7.2 Placing Markers

In addition to covering objects and preventing things crossing the scene, the sur-
veyor also has to place and identify objects in it prior to pushing the button.
Objects may be markers, such as nails, or marked existing points, such as sharp
corners. They are used as control points, necessary for the conversion of range
data to 3D coordinates of object points in a national or local reference system.
The location of the points has to be selected with care and unambiguously indi-
cated to avoid mistakes. Most commonly, four control points are used at the edges
of the scan, although three is enough for geo-referencing; the fourth is redundant
and used for reserve and check. Large objects require scanning from several posi-
tions, and adjacent scans have adequately to overlap. Once a control point has been
identified in one scan, identification in the overlapping scans can be done semi-
automatically.
6.8 Applications 115

6.8 Applications
Laser scanning has been established as a technique for precise surveying of com-
plex objects and scenes and has become a consolidated practice for many surveying
firms and other users. This section discusses a number of applications brought to the
footlight during user conferences and discussed in literature.

6.8.1 3D City Modelling

During the fifth user conference of Leica Geosystems HDS held in San Ramon,
California, autumn 2007, a broad pallet of applications were brought to stage (see
Lemmens, 2008). General Motors Corporation, Detroit, uses 3D as-built models
of its plants all over the world created with TLS, giving engineers and suppliers
instantly access to the data via the web. The plant sector shows a strong rise in using
laser scanning in part driven by the 2006 release of a phase-shift scanner which
can measure up to 500,000 points per second. Highly detailed, photo-textured 3D-
city models of Glasgow have been created by the Glasgow School of Art, a project
funded by the European Union. The data were captured by two scanning surveyors
and the 3D models generated by six, sometimes eight, modellers, over about an 18-
month period, starting in the summer of 2006. Since Glasgow is an old city with
many narrow and curved streets, 3D modelling needs substantially more time than
for a new city like Toronto. The University of Rome, Italy, has developed a 3D model
of the Coliseum and scanned the interior of today’s St. Peter’s Cathedral (heritage
documentation is not so much an industry as a university research activity). TLS can
be even applied to model such complex structures as domes (Fig. 6.7).

6.8.2 Traffic Accidents and Road Safety

The police and the military also benefit from TLS for such applications as traffic-
accident survey and forensic research. California Highway Patrol (CHP) uses TLS
for accident surveying, since 2006 when they bought no fewer than five scanners.
Each accident scene needs to be surveyed quickly, accurately and completely. An
extraordinary road accident took place in the early morning of 29 April 2007 on
an approach to the San Francisco-Oakland Bay Bridge. A tanker truck carrying
over 30,000 litres of gasoline crashed and went up in flames. The heat caused an
overpass to collapse onto an interstate (highway) below. A few minutes later CHP’s
Multidisciplinary Accident Investigation Team was on site and scanned the overpass
from ground to top. With conventional means just five hundred to a thousand points
can be collected with laser scanning, millions. Unsafe positions can be measured
remotely. Also, the road can be reopened sooner and high detail and accuracy allows
not only accident reconstruction but is also useful for civil engineers tasked with
renovating the road construction.
116 6 Terrestrial Laser Scanning

Three-dimensional road capture is not only a feasible measurement method once

an accident has happened but also to study safety of road parts such as crossings in
order to prevent future accidents. TLS can become an important measurement tech-
nique for this type of applications because (inter)national road safety programmes
are gaining growing interest and with these come a need for 3D documentation of
roads for sight analysis and traffic-accident reconstruction studies (Pagounis et al.,
2007). Also crime scenes can be reconstructed by combining laser-scan data taken
after the crime with forensic information. The resulting simulations of the past can
be taken to court and seem to be visually convincing, imitating the realism of film,
and so often convinces juries that this is how it happened.

6.8.3 Deformation and Heritage

The use of TLS for deformation measuring is gaining increasing interest. Monserrat
and Crosetto (2008) developed a new procedure for land deformation monitoring
using a sequence of point clouds taken at different moments in time (multi-
temporal). The method aims at applicability in a wide range of applications and
is based on least squares 3D surface matching as developed at ETH Zurich, by
Professor Gruen and his Group. TLS is also applicable for deformation and load
test measurements of bridges as Lovas et al. (2008) demonstrate, see also (Lovas and
Berényi, 2009). They conducted a test on two Danube bridges, Budapest, one being
a suspension bridge and the other a tied-arch bridge. TLS can measure displace-
ments of cables and pylons not possible by traditional methods. Since the accuracy
is less than for high-precision land surveying instruments, accompanying use of
these instruments remains necessary.
TLS has been successfully applied in the documentation of historic buildings
and archaeological features all over the world because this measurement technique
enables quickly collecting reliable, high-resolution data and saves up to a hun-
dred man-hours over a typical 1-month excavation (Neubauer, 2007). In addition
to documentation the data can be used for creating virtual-reality models, restora-
tion planning or virtual reconstruction and other products attractive for the general
public. Combining terrestrial laser-scanning, magnetic field and electrical resis-
tance data may also reveal anomalies at heritage sites that remain concealed during
fieldwork and in aerial photos, which was demonstrated during preservation of Por-
Bajin fortress, Siberia (Anikushkin and Kotelnikov, 2008). Figure 6.11 shows the
Trimble laser scanner in action to capture the Por-Bajin fortress site. Section 12.2 in
Chapter 12 provides a detailed overview applying TLS in recreating the past.

6.8.4 Other Applications

Foresters can use TLS for determining stand value and log products (Murphy, 2008).
TLS has been successfully applied in a wide variety of civil engineering settings.
6.9 Mobile Mapping 117

Fig. 6.11 Trimble GS200 at work at the Por-Bajin Fortress heritage site, Siberia

6.9 Mobile Mapping

A mobile mapping system (MMS) consists basically of a car, van or boat equipped
with a positioning system – comprising a GNSS receiver integrated with an Inertial
Measurement Unit (IMU) – and laser scanners (Tao and Li, 2007). Digital cameras,
thermal sensors or other geo-data capturing systems may also be mounted on the
roof rack. Figure 6.12 shows an example of a car equipped with GNSS, IMU, laser
scanners and cameras. For inspection of the condition of construction surfaces, such
as road asphalt, ground penetrating radar sensors may be attached to the vehicle
too. Mobile mapping systems are sometimes preceded by the prefix ‘terrestrial’ to
distinguish them from airborne platforms equipped with similar geo-data acquisition
sensors. The positioning system comprising a GNSS receiver integrated with an
Inertial Measurement Unit (IMU) enables to geo-reference the data collected by the
sensors. In urban areas GNSS satellite signals may be blocked by high-rise buildings
and other tall objects (urban canyons). To bridge the time GNSS satellite signals
are blocked attaching odometers to the vehicle wheels is beneficial for improving
the geometric quality of the data. These three redundant positioning technologies –
GNSS, IMU and odometers – enable to obtain accurate positions of the laser and
other sensors while they move around and acquire data. All devices are connected
to a control unit from which the operator steers the capture of 3D coordinates of
millions of points.
To save costs it is important that the time necessary for mounting, installing and
setup of all devices is reduced to minimum. The cost per kilometre of surveying by
means of MMS is falling by the year. Although the diverse manufacturers built a
wide variety of systems, a typical MMS configuration may consist of the following
components (Fig. 6.13 ):
118 6 Terrestrial Laser Scanning

Fig. 6.12 Mobile mapping system; note the odometer at the driving wheel at the rear. At the other
side, not visible, a second odometer is mounted

Fig. 6.13 Mitsubishi MMS-X640 mounted on top of a car. The configuration comprises from left
to right: laser scanner, digital camera, GNSS antenna on top of an IMU, digital camera and laser
scanner
6.9 Mobile Mapping 119

– One GNSS receiver, measuring 10 times per second the position of the car while
being locked to L1 and L2 GPS signals and Glonass signals
– An IMU with a data rate of 100 measurements per second
– Two odometers
– Two digital cameras
– Two laser scanners

After collection of the data in the field the trajectory of the vehicle has to be
computed first and next the point clouds, images and other sensor data have to
be georeferenced. To compute the trajectory all data from GNSS receiver, IMU
and odometers have to be integrated. After georeferencing measurements can be
performed on the 3D point clouds or on overlapping images using a digital pho-
togrammetric workstation to determine the 3D coordinates of, for example, traffic
signs by conventional photogrammetric intersection (Fig. 6.14). The images can also
be used to attach colour to each laser point. Using special software it is possible to
select points in the images and retrieving the coordinates of these point from the
DEM created from the laser points. This is made possible because both the images
and the 3D laser point clouds have been accurately georeferenced.
The coming out of MMS is accelerated because practitioners, such as high-
way managers, are in need of fast and cost-effective data capture systems, while,
in turn, the availability of MMS reshapes the Lidar mapping market allowing the
introduction of new applications and end-customers (El-Sheimy, 2005). The main
application with a high potential of commercial success is mapping of highways and
other roads, necessary for construction, maintenance, safety and environmental pur-
poses. Surveying of roads by traditional means is a time- and resource-consuming
task. Mobile mapping systems provide an enormous time saving in capturing road
networks and their surroundings (Cheng et al., 2008). MMS is able to acquire up
to one billions points per hour together with tens of gigabytes of imagery. Field
trials have shown that accuracies up to two centimetres can be achieved (Sukup
and Sukup, 2010). So, high-quality geometric information of objects on, above and
alongside roads can be extracted from dense point clouds. Data capture while the

Fig. 6.14 Measuring 3D coordinates of a traffic sign from two overlapping images captured during
a mobile mapping survey
120 6 Terrestrial Laser Scanning

vehicle is moving with a speed of 100 km per hour is just the first link in the chain of
providing proper information to users. The huge volumes of data require dedicated
software to manage and analyse the billions of points. To speed up the cumbersome
manual extraction of features much is done for developing software for extracting
road signs, roadsides and the like automatically; research institutes are doing a lot of
work is this realm. To allow access to the data anytime, anywhere by a wide range of
professionals and other users a web-portal consisting of at least a web-based viewer
should be employed.

References
Anikushkin M, Kotelnikov S (2008) 3D-model of Siberian fortress. GIM Int 22(7):13–15
Bacciocchi M, Byham P, Conforti D (2009) Data integration for coastal surveying: combining laser
scanner and bathymetric sensor. GIM Int 23(3):13–15
Bertolotti M (1983) Masers and lasers: an historical approach. Adam Hilger, Bristol
Byham P, Bacciocchi M, Conforti D (2007) Waterside mapping in Italy: integrating 3D laser
scanning and side-scan sonar. GIM Int 21(8):47–49
Cheng W, Hassan T, El-Sheimy N, Lavigne M (2008) Automatic road vector extraction for mobile
mapping systems. Int Arch Photogramm Remote Sens 37(Part B3b):515–521
Colombo L, Marana B (2007) Camera laser scanner: geo-knowledge for preservation of St Maria
Maggiore, Italy. GIM Int 21(8):15–18
Colombo L, Marana B (2010) Terrestrial laser scanning: how it works and what it does. GIM Int
24(12):17–20
Conforti D (2010) Measuring open pit mines: testing static and mobile laser scanners. GIM Int
24(8):17–19
El-Hakim S, Gonzo L, Voltolini F, Girardi S, Rizzi A, Remondino F, Whiting E (2007) Detailed
3D modelling of castles. Int J Archit Comput 5(2):199–220
El-Hakim S, Remondino F, Voltolini F (2008) 3D Modelling of castles: integrating techniques for
detail and photo-realism. GIM Int 22(3):21–25
El-Sheimy N (2005) An overview of mobile mapping systems. FIG Working Week 2005, Cairo,
Egypt, pp 1–24
Heritage GL, Large ARG (eds) (2009) Laser scanning for the environmental sciences. Wiley-
Blackwell, Chichester, West Sussex, UK
Koechner W (1992) Solid-state laser engineering, 3rd edn. Springer, Berlin. ISBN 0-387-53756-2
Lee S, Lee JO, Park HJ, Bae KH (2010) Investigations into the influence of object characteristics
on the quality of terrestrial laser scanner data. KSCE J Civ Eng 14(6):905–913
Lemmens M (2004) 3D Laser mapping. GIM Int 18(12):44–47
Lemmens M (2006) 3D Laser scanner software. GIM int 20(9):49–53
Lemmens M (2007a) Lidar. GIM Int 21(2):11
Lemmens M (2007b) Laser scanning technology challenged: shortcomings in spatial documenta-
tion of heritage sites. GIM Int 21(3):25–29
Lemmens M (2007c) Merging professions. GIM Int 21(8):11
Lemmens M (2007d) Terrestrial Laser Scanners. GIM int 21(9):41–45
Lemmens M (2008) Leica geosystems, HDS: fifth user conference and HQ, San Ramon, USA.
GIM Int 22(2):37–41
Lemmens M (2009) Terrestrial laser scanners. GIM int 23(8):62–67
Lemmens M (2010) Terrestrial laser scanners. GIM Int 24(8):30–33
Lemmens MJPM, van den Heuvel FA (2001) 3D Close-range mapping systems; detailed and
accurate 3D object models from laser. GIM Int 15(1):30–33