The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B4. Beijing 2008
Overall Accuracy = (TP + TN) / Total number of pixels
The Overall accuracy of our building detection system equals to
95.1% which shows the percentage of correctly classified pixels.
5. CONCLUSIONS AND FURTHER REASEARCH
This paper presented an automatic building detection system by
the use of LIDAR point cloud data provided in two individual
files as first pulse and last pulse returns of laser pulse. Our
system is comprised of five steps explained successively in
section 3. In the resulting image, pixels are assigned either
“Building” or “Non-building” labels.
The results of the pixel-by-pixel comparison method used here
proved that our building detection system has made some
improvements in the detection task in comparison with some
previous works. In addition to the Completeness and
Correctness metrics, the evaluated 95.1% Overall accuracy of
our system proves its efficiency and relatively high accuracy.
We hope to make a clear comparison between our method and
the existing ones in future works to claim the capabilities of our
method.
The study area contains a few flat-roofed, distanced buildings
with some high-rise vegetation between them. So the accuracy
of the system is not tested for any other scenes including
complex scenes or dense urban areas. It is also recommended
that the main version of Sohn filtering be used for more
complex scenes and areas with significant topographic change
effects.
Different datasets including buildings with various roof types
are recommended to be included for the assessment of the
system, or any further corrections to the whole algorithm.
REFERENCES
Burrough, P. A. and McDonell, R.A., 1998. Principles of
Geographical Information Systems (Oxford University Press,
New York), p. 190.
Guo, T., Yasuoka, Y., 2002. Snake-based approach for building
extraction from high-resolution satellite images and height data
in urban areas. Proceedings of the 23rd Asian Conference on
Remote Sensing, November 25-29, Kathmandu. 7 pp., on CD-
ROM.
Keqi Zhang, Jianhua Yan, and Shu-Ching Chen. Automatic
Construction of Building Footprints From Airborne LIDAR
Data. IEEE TRANSACTIONS ON GEOSCIENCE AND
REMOTE SENSING, VOL. 44, NO. 9, SEPTEMBER 2006
Lee, D.S., Shan, J., Bethel, J.S., 2003. Class-guided building
extraction from IKONOS imagery. Photogrammetric
Engineering and Remote Sensing 69 (2), 143-150.
Maas, H.-G., Vosselman, G., 1999. Two algorithms for
extracting building models from raw laser altimetry data. ISPRS
Journal of Photogrammetry and Remote Sensing 54 (2-3), 153—
163.
Rottensteiner, F., Trinder, J., Clode, S., Kubik, K., 2005. Using
the Dempster-Shafer method for the fusion of LIDAR data and
multi-spectral images for building detection. Information
Fusion, Volume 6, Issue 4, December 2005, pp 283-300
Schenk, T., Csatho, B., 2002. Fusion of LiDAR data and aerial
imagery for a more complete surface description. International
Archives of Photogrammetry, Remote Sensing and Spatial
Information Sciences 34 (Part 3), 310-317
Sithole, G., Segmentation and Classification of Airborne Laser
Scanner Data (PhD Thesis). Publications on Geodesy 59 NCG
(Netherlands Geodetic Commission) Delft, May 2005
Sohn, G., Dowman, I.J., 2002. Terrain surface reconstruction by
the use of tetrahedron model with the MDL criterion.
International Archives of Photogrammetry, Remote Sensing and
Spatial Information Sciences 34 (Part 3), 336-344.
Sohn, G., Dowman, I., 2007. Data fusion of high-resolution
satellite imagery and LiDAR data for automatic building
extraction. ISPRS Journal of Photogrammetry & Remote
Sensing 62 (2007) pp.43-63, 15 February 2007
Vosselman, G., 1999. Building reconstruction using planar
faces in very high density height data. International Archives of
Photogrammetry and Remote Sensing 32 (Part 2), 87-92.