improved from 6311 sec for about 1,720,000 scanned points to 
2845 sec for about 21,800,000 points. Therefore, the 
considerable improvements in the recognition accuracy and 
computational performance were achieved in this research. 
However, the accuracy slightly degraded in the chemical plant. 
The reason of the degradation was speculated that there are lot 
of points on the straight pipes very close to the non-straight 
pipes, and the situation induced the indispensable error in the 
normal tensor evaluation. 
4. CONCLUSION 
A new algorithm was proposed that could automatically 
recognize a piping system from registered laser scanned points 
of a plant. Normal-based region growing allowed extracting the 
points on piping system. Eigen analysis of the normal tensor 
and cylinder surface fitting allows the algorithm to accurately 
recognize portions of straight pipes. Tracing axes, fitting arc- 
line segments and complementing of segments realized the 
accurate recognition of the position of elbows and junctions and 
their connection relationship. The recognition accuracy was 
verified for large-scale point clouds of actual plants, and the 
considerable improvements in the recognition accuracy and 
computational cost were realized 
However, some normal tensors could not correctly evaluate in 
the straight pipes in very tangled space. This remains to be part 
of our future work. 
References: 
Masuda. H, Tanaka. I, Enomoto. M, 2012, "Reliable Surface Extraction from Point- 
Clouds using Scanner-Dependent Parameters”, Computer-Aided Design and 
Applications, Vol.10, No.2, pp.265-277. 
Rabbani. T, Heuvel. F, 2004, “3D Industrial Reconstruction by Fitting CSG Models to a 
Combination of Images and Point Clouds”, Proceedings of ISPRS The International 
Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol.35, 
No.B3, pp.7-12. 
Aurelien. B, Chaine. R, Marc. C, Thibault. G, Akkouche. S, 2011, "Reconstruction of 
Consistent 3D CAD Models from Point Cloud Data Using A Priori CAD Models", 
Proceedings of ISPRS Workshop Laser Scanning 201 1, submission-26. 
Johnson. A, Hoffman. R, Osborn. J, Hebert. M, 1997, *A system for semi-automatic 
modeling of complex environments", Proceedings of IEEE International Conference on 
Recent Advances in 3-D Digital Imaging and Modeling, pp.213-220. 
Lee. J, Kim. C, Son. H, Kim. C, 2012, *Skeleton-Based 3D Reconstruction of As-Built 
Pipelines from Laser-Scanned Data", Proceedings of the 2012 ASCE International 
Conference on Computing in Civil Engineering, pp.245-252. 
Belton. D, Lichti. D, 2006, "Classification and Segmentation of Terrestrial Laser 
Scanner Point Clouds Using Local Variance Information", The International Archives of 
the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol.36, No.5, 
pp. 44-49 
Rabbani. T, Heuvel. F, Vosselman. G, 2006, "Segmentation of point clouds using 
smoothness constraint", International Archives of the Photogrammetry, Remote Sensing 
and Spatial Information Sciences, Vol. 36, pp. 248-253. 
Vosselman. G, Gorte. B.G.H, Sithole. G, Rabbani. T, 2004, “Recognising Structure in 
Laser Scanner Point Clouds”, International Archives of Photogrammetry, Remote 
Sensing and Spatial Information Sciences, Vol.46, pp.33-38. 
El-Halawany. S, Lichti. D, 2011, “Detection of road poles from mobile terrestrial laser 
Scanner point cloud”, Proc. of 2011 International Workshop on Multi- 
Platform/MultiSensor Remote Sensing and Mapping (M2RSM), pp. 1-6. 
Marshall. A. D, Lukacs. G, Martin. R. R, 2001, “Robust segmentation of primitives from 
range data in the presence of geometric degeneracy”, IEEE Transactions on Pattern 
Analysis and Machine Intelligence, Vol.23, No.3, pp.304-314. 
Kawashima. K, Kanai. S, Date. H, 2011, “Automatic Recognition of a Piping System 
from Large-scale Terrestrial Laser Scan Data”, Proceedings of ISPRS Workshop Laser 
Scanning 2011, submission-20. 
Shakarji. M. C, 1998, “Least-Squares Fitting Algorithms of the NIST Algorithm Testing 
System”, Journal of Research of the National Institute of Standards and Technology, 
Vol.103, No.6, pp.633-641. 
Bauer. U, Polthier. K, 2009, “Generating parametric models of tubes from laser scans”, 
Computer-Aided Design, Vol.41, No.10, pp.719-729. 
  
126 
  
   
14.8 [m] 
  
    
A 
   
(b) Final recognition 
Figure.8 Recognition result of piping system of an oil rig 
Table.1 Accuracy of piping system recognition of an oil rig 
(a) Straight pipe (b) Connecting parts 
Result of automatic recognition 
nition |negative | positive nition 
  
  
  
  
  
  
11.4 [m] 8.6 [m] 
(a) Scanned point cloud (21,880,374 points) 
  
  
  
  
  
  
(b) Final recognition 
Figure.9 Recognition result of piping system 
of a chemical plant 
Table.2 Accuracy of piping system recognition of 
a chemical plant 
(a) Straight pipe (b) Connecting parts 
Result of automatic recognition 
    
recog- nition | negative | positive nition 
Elbow 
o 
> 
= 
Others 
Total 
KEY WO 
ABSTRA 
Registrati 
well as th 
point clou 
transforme 
depth me: 
demonstra 
Since thei 
such as t 
popularity 
Kinect sei 
20-30 fre 
coloured 
Compared 
accuracy 
high data 
make it ar 
environme 
A primary 
Successive 
features, i 
images by 
SIFT (Lo 
correspon 
depth dat 
translatior 
The pairw 
error of ir 
colour sp 
pairwise r 
in the res 
loop closi 
Endres et 
of the sen: 
that is pre 
adjustmen 
simultane 
Loop clos 
a long na 
do not ha 
  
* Corres,