In: Paparoditis N., Pierrot-Deseilligny M., Mallet C.. Tournaire O. (Eds). IAPRS. Vol. XXXVIII. Part ЗА - Saint-Mandé, France. September 1-3, 2010 
Table 1: Object-based evaluation results in percentages (C m = 
completeness, C r = correctness, Qi = quality, M d = multiple de 
tection rate, D 0 = Detection overlap rate, C r d = detection cross 
lap rate and C rr = reference cross-lap rate). 
Scenes 
C m 
a 
Qi 
M d 
D 0 
C r d 
Crr 
Scene 1 
98.6 
97.2 
95.9 
4.1 
5.4 
1.4 
5.7 
Scene 2 
95.2 
95.2 
90.8 
3.1 
3.1 
1.6 
6.5 
Scene 3 
98.3 
92.2 
90.8 
4.5 
9.0 
13.4 
23.3 
Scene 4 
95.1 
95.1 
90.7 
6.1 
18.3 
17.5 
28.7 
Average 
95.9 
94.7 
91.4 
5.1 
12.5 
12.5 
21.7 
Table 2: Pixel-based evaluation results in percentages (C mp = 
completeness, C rp = correctness, Qi v = quality. A oe = area omis 
sion error. A cc = area commission error, Bf = branching factor 
and Mf = miss factor). 
Scenes 
Cmp 
C rp 
Qip 
A oe 
A ce 
Bf 
M f 
Scene 1 
78.5 
89.0 
71.5 
21.6 
10.7 
12.3 
27.5 
Scene 2 
77.7 
87.4 
69.8 
22.3 
12.3 
14.5 
28.8 
Scene 3 
80.5 
91.4 
74.8 
19.5 
8.3 
9.5 
24.3 
Scene 4 
81.4 
85.1 
71.3 
18.6 
14.1 
17.5 
22.9 
Average 
80.4 
87.5 
72.0 
19.7 
12.0 
14.5 
24.6 
6 CONCLUSIONS AND FUTURE WORK 
This paper has proposed an automatic building detection tech 
nique using LIDAR data and multispectral imagery. The initial 
building positions are obtained from the primary building mask 
derived from LIDAR data. The final building positions are ob 
tained by extending their initial positions based on colour infor 
mation, and the two masks ensure the accurate delineation of the 
buildings. In particular, the primary building mask helps separate 
building detections when they are very close to each other and the 
secondary building mask helps to confine the extension of initial 
positions outside a building when the roof and ground have sim 
ilar colour information. Experimental testing has shown that the 
proposed technique can detect rectilinear buildings of different 
shapes with a very high success rate. 
An important observation from the presented results is that object- 
based completeness (detection rate 95.9%) is high when com 
pared to pixel-based completeness (matching overlay 81.4%). How 
ever. the geometric positional accuracy remains relatively poor 
(14.5 pixels) for mapping purposes; although not for applications 
where building detection is the primary goal. This observation 
suggests that the proposed detection technique can be applied 
in city planning, homeland security, disaster (flood or bushfire) 
management and building change detection with high reliability, 
but it is not as yet applicable to cadastral mapping and accurate 
roof plane extraction, both of which require higher pixel-based 
and geometric accuracy. 
REFERENCES 
Awrangjeb, M.. Lu, G., Fraser, C. S. and Ravanbakhsh, M., 2009. 
A fast corner detector based on the chord-to-point distance accu 
mulation technique. In; Proc. Digital Image Computing: Tech 
niques and Applications, Melbourne, Australia, pp. 519-525. 
Awrangjeb. M., Ravanbakhsh. M. and Fraser, C. S., 2010. Auto 
matic detection of residential buildings using lidar data and multi 
spectral imagery. ISPRS Journal of Photogrammetry and Remote 
Sensing. 
BaristaSoftware, 2010. www.baristasoftware.com.au. 
Cheng. L., Gong. J., Chen, X. and Han. P., 2008. Building bound 
ary extraction from high resolution imagery and lidar data. Inter 
national Archives of the Photogrammetry, Remote Sensing and 
Spatial Information Sciences 37(part B3), pp. 693-698. 
Demir, N.. Poli, D. and Baltsavias, E.. 2009. Extraction of build 
ings using images & lidar data and a combination of various 
methods. Int. Archives of the Photogrammetry, Remote Sensing 
and Spatial Information Sciences 38(part 3AV4), pp. 71-76. 
Foody, G., 2002. Status of land cover classification accuracy as 
sessment. Remote Sensing of Environment 80( 1). pp. 185-201. 
Lee, D.. Lee. K. and Lee. S.. 2008. Fusion of lidar and imagery 
for reliable building extraction. Photogram metric Engineering 
and Remote Sensing 74(2). pp. 215-226. 
Lee, D., Shan, J. and Bethel. J., 2003. Class-guided building 
extraction from ikonos imagery. Photogrammetric Engineering 
and Remote Sensing 69(2). pp. 143-150. 
Mayer, H.. 1999. Automatic object extraction from aerial im 
agery' - a survey focusing on buildings. Computer Vision and 
Image Understanding 74(2). pp. 138-149. 
Rottensteiner, F., Trinder. J., Clode, S. and Kubik, K., 2005. Us 
ing the dempstershafer method for the fusion of lidar data and 
multi-spectral images for building detection. Information Fusion 
6(4). pp. 283-300. 
Rottensteiner. F.. Trinder, J.. Clode. S. and Kubik, K., 2007. 
Building detection by fusion of airborne laser scanner data and 
multi-spectral images : Performance evaluation and sensitivity 
analysis. ISPRS Journal of Photogrammetry and Remote Sens 
ing 62(2). pp. 135-149. 
Rutzinger. M., Rottensteiner. F. and Pfeifer. N., 2009. A compari 
son of evaluation techniques for building extraction from airborne 
laser scanning. IEEE Journal of Selected Topics in Applied Earth 
Observations and Remote Sensing 2(1), pp. 11-20. 
Shan, J. and Lee, S.. 2005. Quality of building extraction from 
ikonos imagery. ASCE Journal of Surveying Engineering 131(1), 
pp. 27-32. 
Shufelt, J.. 1999. Performance evaluation and analysis of monoc 
ular building extraction from aerial imagery. IEEE Trans, on Pat 
tern Analysis and Machine Intelligence 21(4), pp. 311-326. 
Sohn, G. and Dowman. I., 2007. Data fusion of high-resolution 
satellite imagery and lidar data for automatic building extraction. 
ISPRS Journal of Photogrammetry and Remote Sensing 62(1), 
pp. 43-63. 
Song. W. and Haithcoat. T., 2005. Development of comprehen 
sive accuracy assessment indexes for building footprint extrac 
tion. IEEE Transactions on Geoscience and Remote Sensing 
43(2), pp. 402-A04. 
Sun, J., Lin. Y., Kang, S. and Shunt, H., 2005. Symmetric stereo 
matching for occlusion handling. In: Proc. IEEE Conference 
on Computer Vision and Pattern Recognition, Vol. 2, San Diego, 
CA, USA, pp. 399-406. 
Vu. T., Yamazaki, F. and Matsuoka. M., 2009. Multi-scale solu 
tion for building extraction from lidar and image data. Interna 
tional Journal of Applied Earth Observation and Geoinformation 
11(4), pp. 281-289. 
Yong, L. and Huayi, W., 2008. Adaptive building edge detection 
by combining lidar data and aerial images. International Archives 
of the Photogrammetry', Remote Sensing and Spatial Information 
Sciences 37(pail Bl), pp. 197-202.