- 103 - 
ii. Laser scanning is far more accurate and many times faster 
than digital photogrammetry in providing surface detail of 
stones in archaeological and architectural applications than 
digital photogrammetry. Both the data acquisition and the 
restitution phases in laser scanning are much faster and 
more accurate than the comparable process in digital photo 
grammetry (or conventional photogrammetry for that mat 
ter). 
These findings suggested to 3DID that the best way forward 
would be to develop a hybrid approach wherein the outlines are 
produced using digital photogrammetry and then these could be 
filled in with much richer data obtained through laser scanning. 
4. Thealasermetry methodology 
4.1 Identifying client needs and functional specifications 
The hybrid approach combining theodolite, laser scanning and 
digital photogrammetry was developed to try and get the best of 
all worlds into one integrated approach. From the above obser 
vations in the field and the post-processing laboratory it is im 
portant to note that, as ever, the surveyor must be careful to 
identify exactly what the client requires i.e. what the ultimate 
functional specifications of the survey are to be. As has been 
seen, some surveys would require maximum detail about sur 
face texture of the stone, thus mandating the use of a certain 
type of laser scanner. Other surveys would not give such prior 
ity to surface texture and, especially in the case of large archi 
tectural sites, would mandate the use of another type of laser 
scanner. 
4.2 Two key elements of Thealasermetry: common reference 
points & data transfer capability 
Early on in the process 3DID identified two key elements re 
quired to successfully marry laser scanning and digital photo 
grammetry in a way which would enable the one to best com 
plement the other: 
i. The approach must be able to use as many com 
mon reference points as possible which can be 
recognized by all of the three technologies em 
ployed: theodolite, photgrammetry and laser 
scanner. 
ii. The team must be able to transfer data produced 
by one technology to the other 
4.3 A step-by-step process for the hybrid approach 
1) The first and most crucial step is the establishment and use 
of the same reference points for the laser scanner and the 
theodolite total station, so that common co-ordinates will tie 
different scans together. The theodolite total station co 
ordinates are used as the binding factor for the consolidation 
of all the data gathered from the three systems. Therefore, 
these co-ordinates are used as the reference points for the 
control markers used for the photogrammetry and laser 
scanning during the data acquisition phase. 
2) Data acquisition by all three systems, (theodolite, laser 
scanner and photogrammetry), is followed by the second 
step which is that of consolidating data sets from two differ 
ent sources. The 3DID documentation specialist currently 
has a choice: either to combine the photogrammetry co 
ordinates system into the laser restitution software (3D Ip- 
sos) or inserti the laser co-ordinates system into the photo 
grammetry software (DIAP). 
3) The third step is that of creating a triangulated surface out of 
the point cloud data obtained as a result of the laser scan. 
This process, known as meshing, is a representation of a 
surface, where points are linked together to form a solid sur 
face. Both 3DIPSOS and DIAP have meshing functions 
with different levels of automation available. 
4) The fourth step is that of using the photogrammetry soft 
ware (DIAP) to create outlines of different stones using the 
stereo paired digital photographs. 
5) The fifth step is to import lines drawn with photogrammetry 
software to 3D IPSOS (the laser scanner restitution soft 
ware) - outlines are easily identifiable by using the stereo 
pairs. The outlines imported/exported from the photo 
grammetry software enable the user to recognise the objects 
true line. With the outline, errant triangles created at the 
meshing stage can be deleted. 
6) The sixth step is texture mapping (if required at specifica 
tion stage). The computer model is rendered to give it a re 
alistic and solid look. At this stage, the documentation spe 
cialist can also choose digital pictures taken by the laser 
scanner’s built-in digital camera or digital pictures taken at 
a higher resolution during the photogrammetry data acqui 
sition process in order to use these to paste on to his com 
puter model in order to enhance the realistic appearance of 
the final result. 
5. Experimental Results and Conclusions: 
From a hybrid approach to hybrid machines 
In summary therefore, the theodolite total station’s co-ordinates 
are taken as the first ‘viewpoint’. Using this system, the laser 
scan viewpoints can be consolidated within the total station’s 
closed-network. Using the common set of co-ordinates to resti 
tute the photogrammetric stereopairs, very accurate outline 
drawings can be obtained. The results obtained from the differ 
ent processes are merged into one system. This allows the tri 
angulation of the laser scanning points to be modelled within 
the vector outlines estimated using the photogrammetric stere 
opairs.The same photographs used for the photogrammetry can 
be directly utilized for texture mapping of the integrated data. 
Having the photographs directly linked to the total station, en 
sures that this process is completed with the maximum accuracy 
and ease possible. Overall, the expected object precision can 
achieve +/-lmm. Such a system offers an important archive 
which can enable closer monitoring of these sites 
The above process has meant that 3DID can now save more 
than 50% of the time normally spent on back-office post 
processing tasks on digital photogrammetry since this aspect 
can now be restricted to main outline drawing whereas the in 
ternal detail is filled up much faster and to a degree of detail 
hitherto undreamt of by using the data obtained from the laser 
scanner. Using a laser scanner as part of a hybrid approach has 
therefore truly complemented photogrammetry but the level of 
development of the technology to date does not enable laser 
scanning to replace photogrammetry completely due to the in 
accuracies indicated in various sections above.