etry 
entation 
ie most 
id close 
'e more 
ons via 
system 
era). A 
data set 
M 2.5D 
it in the 
rawings 
enerally 
sources 
ditional 
of data 
jS), and 
uctions, 
ey of a 
ntrolled 
ults are 
1entally 
able to 
s based 
ng and 
graphic 
al local 
lologies 
ling the 
imon (o 
that are 
ntations 
empt to 
cessary 
ighlight 
scanner 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B5. Istanbul 2004 
  
Laser scanning and photogrammetry 
Laser scanning has already shown its outstanding advantages in 
acquiring 3D information on an objects surface in many 
different applications within the past few years. For laser 
scanning, a highly collimated laser beam is scanned over a 
predefined solid angle in a regular scan pattern. While scanning, 
the distance to the object is measured by measuring the time of 
flight of the laser signal with high precision. Different 
commercial systems are available with a broad range of 
specifications. The specifications differ in measurement range, 
field-of-view, measurement accuracy, data acquisition speed, 
robustness, compactness, and transportability. The primary 
output delivered by a scanning laser system is a point cloud 
representing a sampled replica of the objects surface. The point 
cloud is composed usually of a very large number of points or 
vertices and, for most of the systems, each vertex corresponds to 
a single laser range measurement. As for many applications the 
user is not only interested in geometrical information, but also 
in additional information on the object's surface, the point cloud 
is frequently complemented by additional vertex descriptors 
containing information on, e.g., surface reflectivity or surface 
color. Almost all laser scanners provide, beside the geometry 
data, also information of the signal strength of the echo signal, 
commonly addressed as intensity data. Some laser sensors 
provide with every laser measurement also color information by 
converting the ambient light in the direction of the laser beam 
into an RGB (red-green-blue) triple. The geometrical data and 
the additional vertex descriptors are acquired synchronously and 
sequentially and the spatial resolution of the additional data can 
thus not be higher. In order to have texturing data with a higher 
resolution than the laser data, high resolution digital cameras 
can be used additionally. 
  
  
  
In principle, texturing 3D models generated from laser scan data 
with image data is well-established and many of the 3D data 
537 
processing packages provide at least some means for texturing 
the surface of a 3D model. However, using images of a camera 
without prior knowledge of its position and orientation requires, 
for example, manual definition of tie points in both the scan 
data and the image to calculate the image parameters. 
Integrating a high-resolution calibrated camera into a laser 
scanning system provides a very efficient, convenient, and 
powerful system for automatically generating accurately 
textured high-resolution 3D models. This combination forms a 
hybrid sensors which is composed of a high-performance long- 
range laser scanner with a wide field-of-view and a calibrated 
high-resolution digital camera firmly mounted to the scanning 
head of the laser scanners. As for every image taken with the 
camera the position and orientation of the camera is measured 
with high accuracy within the scanners own coordinate system, 
scan data and image data can be combined in a straightforward 
way without the need of user interaction. 
  
fig.3 LMS-Z 360i and phototheodolite. 
Utilization of this new type of hybrid data in the field of 
Cultural Heritage (particularly architectural and archaeological 
field) can be tackled according to a reductive approach, namely 
attempting to obtain the traditional products more rapidly and 
thus more economically, or from a propositional approach, 
attempting to generate new tools for description and 
representation of complex forms. This challenge involves not 
only surveyors but also other professionals who base their work 
on the results of metric survey, such as restorers, structural 
engineers, and historians. 
In order to satisfy the demands of detailed feature extraction, 
e.g., on ancient walls, RIEGL offers now a unique solution by 
combining laser scanning and close range photogrammetry, thus 
making use of the respective advantages of both. The system 
used for the experimental investigation is a RIEGL LMS-Z360i 
laser scanner with a digital camera, model Nikon D100. It is a 
portable rugged terrestrial sensor, intended for the fast 
acquisition of high-quality 3D images even under difficult 
environmental conditions (Riegl 2004). 
The crucial idea behind this combination is not to regard those 
as competitive but complementary technologies: The most 
impressive result of laser scanning is the definition of surfaces, 
whereas the strength of photogrammetry lies in its capability of 
recognizing edges. The accuracy of the details is caused in the 
high pixel density of the photo camera, the accuracy of the 
whole scene comes from the laser scanner. 
The secondary results derived from the hybrid data are a 
coloured point cloud, a decimated mesh textured with the high 
resolution images, an orthophoto and the possibility to digitize 
the data in the so-called mono plot method.