CMRT09: Object Extraction for 3D City Models, Road Databases and Traffic Monitoring - Concepts, Algorithms, and Evaluation
dataset generally has better performance than the first one in
terms of vehicle motion classification, which has shown that
one-path LiDAR data could be more appropriate for our task
than co-registered data of multiple strips, despite that the point
density of combined dataset would be higher. Moreover, the
superior performance may trace back to the applied extraction
strategy of direct 3D segmentation on LiDAR point clouds
other than 2D analysis approach.
Once the motion status of extracted vehicles is determined, the
velocity of moving vehicles can be inferred under the
precondition that the true vehicle size is known. According to
results presented here, it is easy to empirically give such
performance summery that the vehicle motion indication as
well as estimation from ALS data would fairly depend on
certain factors, such as point density, distribution spacing
between every two vehicles, relative motion direction to the
flight direction, absolute velocity of vehicle, and vehicle size.
The accurate impacts of single factors on motion analysis
results have to be further obtained by quantitative analysis with
great amount of test data
Traffic analysis could quite benefit from some distinctive
operational conditions of LiDAR sensor, in comparison to
optical camera. It is an active sensor less weather dependent;
for example, it can cope with haze, fog and volume-scattering
objects to some extent, working night too. Furthermore, scene
complexity poses an additional difficulty for the optical
imagery: dense urban areas, long and strong shadows,
occlusions, etc., can severely impair the vehicle extraction
performance.
4. CONCLUSION
Overall, a progressive scheme consisting of the vehicle
extraction step followed by motion status classification is
presented in this work attempting to automatically characterize
the traffic scenario in urban areas. Based on single vehicle
instances extracted by an approach combining context
exploitation with 3D segmentation, the binary motion status of
them is determined by shape analysis and classification. As
indicted by the results derived from real ALS data commonly
used for city mapping and modeling, traffic analysis by airborne
LiDAR offers great potential to support the short/mid-term
acquisition of statistical traffic data for a given road network in
urban areas in despite of higher false alarm rates. Nevertheless,
numerous potential improvements of the schemes have to be
developed in future, in order to deal with main obstacles to
LiDAR traffic characterization, especially regarding velocity
estimation, such as low point density, unknown vehicle size and
unstable laser reflection properties of vehicle surface.
REFERENCES
Fletcher, P.T., Conglin, L. and Joshi, S., 2003. Statistics of
shape via principal geodesic analysis on Lie groups. Computer
Vision and Pattern Recognition, 2003. Proceedings. 2003 IEEE
Computer Society Conference on, pp. I-95-I-101 vol.l.
Hinz, S., Bamler, R. and Stilla, U. (Editors), 2006. Theme issue
"Airborne and spacebome traffic monitoring". ISPRS Journal of
Photogrammetry and Remote Sensing, 61(3/4), 135-278 pp.
Hinz, S., Lenhart, D. and Leitloff, J., 2008. Traffic extraction
and characterisation from optical remote sensing data. The
Photogrammetric Record, 23(124): 424-440.
Jarvis, R.A., 1977. Computing the shape hull of points in the
plane, IEEE Computing Society Conference on Pattern
Recognition and Image Processing, New York, pp. 231-241.
Jin, X. and Davis, C.H., 2007. Vehicle detection from high-
resolution satellite imagery using morphological shared-weight
neural networks. Image and Vision Computing, 25(9): 1422-
1431.
Meyer, F., Hinz, S., Laika, A., Weihing, D. and Bamler, R.,
2006. Performance analysis of the TerraSAR-X Traffic
monitoring concept. ISPRS Journal of Photogrammetry and
Remote Sensing, 61(3-4): 225-242.
Reinartz, P., Lachaise, M., Schmeer, E., Krauss, T. and Runge,
H., 2006. Traffic monitoring with serial images from airborne
cameras. ISPRS Journal of Photogrammetry and Remote
Sensing, 61(3-4): 149-158.
Rosenbaum, D., Kurz, F., Thomas, U., Suri, S. and Reinartz, P.,
2008. Towards automatic near real-time traffic monitoring with
an airborne wide angle camera system. European Transport
Research Review.
Rossmann, W., 2002. ‘Lie Groups: An introduction through
linear groups. Oxford University Press.
Runge, H. et al., 2007. Space borne SAR traffic monitoring,
Proceedings, International Radar Symposium, Cologne, pp. 5.
Sharma, G., Merry, C.J., Goel, P. and McCord, M., 2006.
Vehicle detection in 1-m resolution satellite and airborne
imagery. International Journal of Remote Sensing, 27(4): 779 -
797.
Stilla, U., Rottensteiner, F. and Hinz, S. (Editors), 2005. Object
extraction for 3D city models, road databases, and traffic
monitoring— concepts, algorithms, and evaluation (CMRT05)
International Archives of Photogrammetry, Remote Sensing and
Spatial Information Sciences, 36(3/W24), 196 pp.
Toth, C.K. and Grejner-Brzezinska, D., 2006. Extracting
dynamic spatial data from airborne imaging sensors to support
traffic flow estimation. ISPRS Journal of Photogrammetry and
Remote Sensing, 61(3-4): 137-148.
Yao, W., Hinz, S. and Stilla, U., 2008a. Automatic vehicle
extraction from airborne LiDAR data of urban areas using
morphological reconstruction, Proceedings of 5th IAPR
Workshop on Pattern Recognition in Remote Sensing
(PRRS08), Tampa, USA, pp. 1-4.
Yao, W., Hinz, S. and Stilla, U., 2008b. Traffic monitoring
from airborne LIDAR - Feasibility, simulation and analysis,
XXI Congress, Proceedings. International Archives of
Photogrammetry, Remote Sensing and Spatial Geoinformation
Sciences, Beijing, China, pp. Vol 37(B3B):593-598.
Yao, W., Hinz, S. and Stilla, U., 2009. Object extraction based
on 3d-segmentation of LiDAR data by combining mean shift
and normalized cuts: two examples from urban areas, 2009
Urban Remote Sensing Joint event: URBAN 2009 - URS 2009.
Yarlagadda, P., Ozcanli, O. and Mundy, J., 2008. Lie group
distance based generic 3-d vehicle classification, Pattern
Recognition, 2008. ICPR 2008. 19th International Conference
on, pp. 1-4.
Zhao, T. and Nevatia, R., 2003. Car detection in low resolution
aerial images. Image and Vision Computing, 21(8): 693-703.
40
