International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B4, 2012 
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia 
  
Figure 6. Figure shows 3D model of area 2 after applying the 
approach. 
  
Figure 7. Figure shows the isometric view from the area 2. 
5. CONCLUSION 
A new approach for developing a 3D model has been 
mentioned. The approach extracts objects from the point clouds 
and registered on the image, for assisting in object detection and 
extraction from the image. This process is called reverse 
registration. The reverse registration enables the algorithm to 
detect and extract object from the image automatically. Then 
extracted objects from the image will be transformed to a 3D 
space and developed a 3D model. All pixels of the interested 
and extracted objects from the image will be converted to points 
and registered on a developed DTM from point clouds before 
transformed to the 3D space. For extracting object from point 
clouds, the curvature of surface of objects will be assessed. If 
the curvature is changing rapidly, the algorithm will recognise 
the object. In addition of assessment of the curvature, the 
algorithm will assess the extracted objects with a defined 
knowledge which are initially introduced to the algorithm. The 
evaluation of the extracted object confirms that the operator is 
very well able to detect and extract the objects from point 
clouds. The reverse registration improves to detect and extract 
the object from the image. Usually filtering and segmentation 
are common methods for extracting an object from the image, 
but reverse registration assures the process to extract the 
interested object correctly. Indeed the reverse registration is an 
integration process for object detection from the image. The 
application of the approach is very versatile and can be used in 
different purposes; for example, the approach can be used in 
planning and engineering, in analysing objects, in developing 
3D GIS, in medical application, and emergency management. 
6. REFERENCES 
Bae, K.H., Belton, D., Lichti, D., 2009, A Closed-Form 
Expression of the Positional Uncertainty for 3D Point Clouds, 
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND 
MACHINE INTELLIGENCE, Vol 31, No 4, 577-590 
Brahim, N., Güeriot, D., Daniel, S., Solaiman, B., 2010, 3D 
reconstruction of underwater scenes using image sequences 
from acoustic camera, OCEANS IEEE, 1-8 
Chen, Y.W., Kohatsu, T., 2007, 3D Image Reconstruction from 
Limited Projections by Simulated Annealing, Second 
International conference on Innovative Computing Information 
and Control, 456-456 
Dammann, J., Redman, B., Ruff, W., 2006, 3D Image 
Reconstruction and Range-Doppler Tracking with Chirped AM 
Ladar Data, 35th Applied Imagery and Pattern Recognition 
Workshop, 4-4 
Homainejad, A.S., 2011a, A novel approach for constructing a 
3D model based on registering a mono image on a 3D model, 
applicable in Digital Earth, International Journal of Digital 
Earth 
Homainejad, A.S., 2011b, Image registration on image splitting 
and pixel registration, Registration Quality-Towards Integration 
of Laser Scanning and Photogrammetry, Chief Editor: Petri 
Rónnholm, EuroSDR, Official Pblication No. 59, 79-102 
Homainejad, A.S., 2010, An invented approach in image 
registration “new era in photogrammetry”, ISPRS TC VII 
Symposium — 100 Years ISPRS, Vol. XXXVIII, Part 7B, 299- 
303 
Kirchhof, M., Jutzi, B., Stilla, U., 2008, Iterative processing of 
laser scanning data by full waveform analysis, ISPRS Journal of 
Photogrammetry & Remote Sensing 63, 99-114 
Qian, L., 2010, Image 3D Reconstruction System for Indoor 
Environment Based on JAVA 3D, 2nd International Conference 
on Signal Processing Systems (ICSPS), V3-789 - V3-792 
Silvan-Cardenas, J. L., Wang, L., 2006, A multi-resolution 
approach for filtering LIDAR altimetry data, ISPRS Journal of 
Photogrammetry & Remote Sensing, Vol 61, 11-22 
Sportouche, H., Tupin, H., 2009, Leonard DENISE, Building 
Extraction and 3D Reconstruction in Urban Areas from High- 
Resolution Optical and SAR Imagery, Urban Remote Sensing 
Event, 1-11 
Taubin, G., 1995, ESTIMATING THE TENSOR OF 
CURVATURE OF A SURFACE FROM A POLYHEDRAL 
APPROXIMATION, Proceeding of IEEE International 
Conference on Computer Vision, 902-907 
Tupin, F., Roux, M., 2003, Detection of building outlines based 
on the fusion of SAR and optical features, ISPRS Journal of 
Photogrammetry & Remote Sensing 1267, 1-12 
Wei, W., Guorong, W., Hua, C., 2009, 3D Reconstruction of a 
Femur Shaft using a Model and Two 2D X-ray Images, 4" 
international conference on computer science and education, 
ICCSE, 720-722 
Zhang, Z.Y., Tsui, H. T., 1998, 3D Reconstruction from a 
Single View of an Object and Its Image in a Plane Mirror, 
Fourteenth International Conference on Pattern Recognition, 
Vol 2, 1174-1176 
ACKNOWLEGMENT 
Hereby, expressing our thanks to ISPRS commission III, 
working group III/4, specially thanks will be extended to Franz 
Rottensteiner, Caroline Baillard, Gunho Sohn, and Markus 
Gerke, for finding and supporting of ISPRS test project on 
urban classification and 3D building reconstruction and 
Providing data. 
184 
pad p bd Kee Sw aua Ru UI 
e (0) "oM oM) M—— 
dt B^ N22 € r3 fn tz 
E--] 
th