have a 0.5m pixel error, the vehicles’ ground positions will have 
Im error. This paper uses centre of mass of the detected 
vehicles as the vehicles’ image positions. As suggested by 
Xiong and Zhang (2008), a sub-pixel segmentation method will 
improve the accuracy of vehicles’ image positions. 
Furthermore, in this paper, only MS-1 and MS-2 images are 
used to detect two ground positions of a moving vehicle. An 
additional ground position of a moving vehicle can be computed 
from the Pan image which will provide redundant data to find 
and correct the vehicle’s speed. In addition to this, the RPC 
models provided by satellites have a positioning error (Xiong 
and Zhang, 2008) which propagated to the vehicle’s ground 
position. Therefore, sensor refinement can further improve the 
accuracy of vehicle speed computation. 
4. CONCLUSIONS 
A new methodology for automatic moving vehicle detection and 
moving vehicle information extraction from a single pass 
WorldView-2 satellite is presented. This includes two major 
components: (1) A new automatic approach to detect moving 
vehicles from MS-1 and MS-2 images in which there is no need 
to extract roads prior to the vehicle detection; (2) A method to 
extract moving vehicle information (position, speed, and 
direction). The experimental results demonstrate that this 
technique can automatically extract moving vehicles’ 
information from one pass WorldView-2 imagery. Therefore, 
this technique makes it feasible to use WorldView-2 imagery 
for traffic applications on an operational basis. Thus, this 
technique potentially offers a cost effective way to extract 
moving vehicles’ information for traffic management and 
planning. 
Although the developed technique has achieved a fair level of 
accuracy, still there is potential for further improvements such 
as: (1) Automatic road extraction from Pan image prior to 
moving vehicle extraction; (2) Vehicle detection from Pan 
images; (3) An improvement in calculation of vehicles’ image 
coordinates. These improvements will be the part of future 
research. 
REFERENCES 
Coppin, P. Jonckheere, I., Nackaerts, K., and Muys, B., 2004. 
Digital change detection methods in ecosystem monitoring: A 
review. Int. J. Remote Sens., vol. 25, no. 9, pp. 1565-1596. 
Freund, Y., and Schapire, R. E., 1997. A decision-theoretic 
generation of on-line learning and an application to boosting. 
Journal of Computer and System Sciences, vol. 55, no. 1, pp. 
119-139. 
Freund, Y., and Schapire, R. E., 1999, A short introduction to 
boosting. Journal of Japanese Society for artificial 
Intelligence, vol. 14, no. 5, pp. 771-780. 
Friedman, J., Hastie, T., and Tibshirani, R., 2000. Additive 
logistic regression: A statistical view of boosting. Ann. Stat., 
vol. 28, no. 2, pp. 337-407. 
Gerhardinger, A., Ehrlich, D. and Pesaresi, M., 2005. Vehicle 
detection from very high resolution satellite imagery. /nt. Arch. 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012 
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia 
Photogrammetric & Remote Sensing Spatial Information 
Science, vol. 36, no. Part 3/W24, pp. 83-88. 
Jin,X., and Davis,C. H., 2007. Vehicle detection from high- 
resolution satellite imagery using morphological shared-weight 
neural networks. /mage and Vision Computing, vol. 25, no. 9, 
pp. 1422-1431. 
Leitloff, J., Hinz, S., and Stilla, U., 2010. Vehicle Detection in 
Very High Resolution Satellite Images of City Areas. 
Geoscience and Remote Sensing, IEEE Transactions, vol.48, 
no.7, pp.2795-2806. 
Liu, W., Yamazaki, F., Vu, T. T., 2010. Automated Vehicle 
Extraction and Speed Determination From QuickBird Satellite 
Images. Selected Topics in Applied Earth Observations and 
Remote Sensing, IEEE Journal of , vol.PP, no.99, pp.1-8, 0. 
Otsu, N., 1979. A threshold selection method from gray-level 
histograms, /EEE Transaction. Systems, Man and Cybernetics, 
vol. SMC-9, no. 1, pp. 62-66, Jan. 
Tao, J., and Yu, Wenxian.,, 2011. A Preliminary Study on 
Imaging Time Difference among Bands of Worldview-2 and Its 
Potential Applications. 
http://dgl.us.neolane.net/res/img/155c93a0f08ede72818d5a6d17 
c8037c.pdf(Jan 28, 2012) 
Tronin, A. A, 2006. Remote sensing and earthquakes: A review, 
Phys. Chem. Earth, vol. 31, no. 4—9, pp. 138-142. 
Xiong Z., and Zhang,Y., 2008. An initial study on vehicle 
information extraction from single pass of satellite QuickBird 
imagery, Photogrammetric & Remote Sensing, vol. 74, no. 11, 
pp. 1401-1411. 
Zheng H., and Li, L., 2007. An artificial immune approach for 
vehicle detection from high resolution space imagery. 
International Journal in Conputer Science and Network, vol. 7, 
no. 2, pp. 67-72. 
Zheng, H.,Pan, L., and Li, L., 2006. A morphological neural 
network approach for vehicle detection from high resolution 
satellite imagery. in Proc. Int. Conf. Neural Information 
Processing, vol. 4233, Lecture Notes in Computer Science, I. 
King, J. Wang, L. Chan, and D. L. Wang, Eds. New York: 
Springer-Verlag, pp. 99-106. 
ACKNOWLEDGEMENT 
This research was funded by the Canada Research Chairs 
Program. WorldView-2 image was provided by DigitalGlobe® 
Inc. to Bahram Salehi (University of New Brunswick).