CMRT09: Object Extraction for 3D City Models, Road Databases and Traffic Monitoring - Concepts, Algorithms, and Evaluation
Table 4. Error matrix and quality factors of boost k-means
clustering applied to second dataset.
Error Matrix
Reference Map
Building
Tree
Ground
(A
Building
61027
212
1393
3
c/}
Tree
428
10701
1178
C4
Ground
7224
2824
69667
Producer Accuracy
88.9%
77.9%
96.45
Producer Accuracy
97.4%
86.9%
87.4%
Overal Accuracy
91.4%
K-factor
0.850
4. SUMMARY
In this research a boost clustering methodology was applied on
two datasets of LiDAR data in an urban area. The proposed
method is a multiple clustering method based on the iterative
application of a basic clustering algorithm. We evaluated this
algorithm using two datasets, to investigate if this algorithm can
lead to improved quality and robustness of performance. For the
quality analysis of data clustering we used Some quality
analysis factors such as produces, user and overall accuracy
between the true labels and the labels returned by the clustering
algorithms as the quality assessment measure. The experimental
results on LiDAR datasets have shown that boost clustering
algorithm can lead to better results compared to the solution
obtained from the basic algorithm. The usefulness of the two
feature channels Gradient Filtered NDDI and Opening of Last
Pulse Range image for separating vegetation region with 3D
extend and building regions from background has been also
shown by the experiments.
There are also several directions for future work in this area.
The most important is to determine the optimal number of
clusters existing in the dataset. Other interesting future research
topics concern the definition of best features of LiDAR data for
data clustering and also using digital aerial and intensity images
as well as the experimentation with other types of basic
clustering algorithms and comparing the results of boost
clustering with other strong clustering methods such as fuzzy k-
means and neural networks or other multiple clustering based
approaches.
REFERENCES
Alharthy,A., Bethel, J., 2002. Heuristic filtering and 3D feature
extraction from LiDAR data. 1APRS, Graz, Austria, vol.
XXXIII, pp. 29-35.
Bezdek, J., Pal, S., 1992. Fuzzy Models for Pattern
Recognition: Methods that Search for Structures in Data. IEEE
Press, New York, NY. 539 pages.
Dulyakam, P., Rangsanseri, Y., 2001. Fuzzy c-means clustering
using spatial information with application to remote sensing. In:
22nd Asian Conference on Remote Sensing, Singapore, pp. 5-9.
Freund, Y., Schapire, Y., 1996. Experiments with a new
boosting algorithm. In: Proceedings of the Thirteenth
International Conference on Machine Learning (ICML), pp:
148-156.
Freund, Y., Schapire, E., 1999. A short introduction to boosting.
Journal of Japanese Society for Artificial Intelligence, 14(5),
pp. 771-780.
Frossyniotis , D., Likas, A., Stafylopatis, A., 2004. A clustering
method based on boosting. Pattern Recognition Letters, 25, pp.
641-654.
Gonzalez R. C., Woods, R. E. 2006. Digital Image Processing
(3rd Edition), Prentice-Hall, Inc., Upper Saddle River, NJ,
USA.
Jain, A.K., Dubes, R.C., 1988. Algorithms for Clustering Data,
Prentice Hall, New Jersey.
Jain, A. K., Murty, M. N., Flynn, P.J., 1999. Data clustering: a
review. ACM Computing Surveys, 31 (3). pp. 264-323
Kearns, M., Valiant. L. G., 1998. Learning boolean formulae or
finite automata is as hard as factoring. Technical Report TR-14-
88, Harvard University Aiken Computation Laboratory.
Kuncheva, L. I., 2004. Combining Pattern Classifiers, Methods
and Algorithms. John Wiley & Sons, Inc., Hoboken, New
Jersey, pp. 251-253.
Li, X., Wang, L., Sung, E., 2004. Improving adaBoost for
classification on small training sample sets with active
learning, In: The Sixth Asian Conference on Computer Vision
(ACCV), Korea.
Liu, C., Frazier, P., Kumar, L., 2007. Comparative assessment
of the measures of thematic classification accuracy. Remote
Sensing of Environment, 107, pp. 606-616.
Liu, Y., Jin, R., Jain, A. K., 2007. BoostCluster: boosting
clustering by pairwise constraints. In: KDD'07, the 13th
International Conference on Knowledge Discovery and Data
Mining, San Jose, California, USA. pp: 450-459.
Pedrycz, W., 1997. Fuzzy clustering with partial supervision.
In: IEEE Transactions on Systems, Man, and Cybernetics, Part
B: Cybernetics, 27(5), pp. 787-795.
Saffari, A., Bischof, H., 2007. Clustering in a boosting
framework. In: Proceedings of Computer Vision Winter
Workshop, St. Lambrecht, Austria.
Zhong, S., Ghosh, D., 2003. A Unified framework for model-
based clustering. Journal of Machine Learning Research. 4, pp.
1001-1037.
64
