Distribution free decomposition of multivariate data

  • Dorin Comaniciu
  • Peter Meer
Statistical Classification Techniques
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1451)


A practical approach to nonparametric cluster analysis of large data sets is presented. The number of clusters and the cluster centers are derived by applying the mean shift procedure on a reduced set of points randomly selected from the data. The cluster boundaries are delineated using a k-nearest neighbor technique. The resulting algorithm is stable and efficient, allowing the cluster decomposition of a 10000 point data set in only a few seconds. Complex clustering examples and applications are discussed.


Cluster Center Kernel Density Estimate Mean Integrate Square Error Density Constraint Epanechnikov Kernel 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Y. Cheng, “Mean Shift, Mode Seeking, and Clustering”, IEEE Trans. Pattern Anal. Machine Intell., vol. 17, 790–799, 1995.CrossRefGoogle Scholar
  2. 2.
    D. Comaniciu, P. Meer, “Robust Analysis of Feature Spaces: Color Image Segmentation”, Proc. IEEE Conf. on Comp. Vis. and Pattern Recognition, Puerto Rico, 750–755, 1997.Google Scholar
  3. 3.
    J.H. Friedman, J.L. Bentley, R.A. Finkel, “An Algorithm for Finding Best Matches in Logarithmic Expected Time”, ACM Trans. Mathematical Software, vol. 3, 209–226, 1977.CrossRefGoogle Scholar
  4. 4.
    K. Fukunaga, L.D. Hostetler, “The Estimation of the Gradient of a Density Function, with Applications in Pattern Recognition”, IEEE Trans. Info. Theory, vol. IT-21, 32–40, 1975.CrossRefGoogle Scholar
  5. 5.
    K. Fukunaga, Introduction to Statistical Pattern Recognition, Boston: Academic Press, 1990.Google Scholar
  6. 6.
    J.A. Garcia, J.F. Valdivia, F.J. Cortijo, and R. Molina, “A Dynamic Approach for Clustering Data”, Signal Processing, vol. 44, 181–196, 1995.CrossRefGoogle Scholar
  7. 7.
    M. Herbin, N. Bonnet, P. Vautrot, “A Clustering Method Based on the Estimation of the Probability Density Function and on the Skeleton by Influence Zones”, Pattern Recognition Letters, vol. 17, 1141–1150, 1996.CrossRefGoogle Scholar
  8. 8.
    A.K. Jain, R.C. Dubes, Algorithms for Clustering Data, Englewood Cliff, NJ: Prentice Hall, 1988.Google Scholar
  9. 9.
    S.A. Nene, S.K. Nayar, “A Simple Algorithm for Nearest Neighbor Search in High Dimensions”, IEEE Trans. Pattern Anal. Machine Intell., vol. 19, 989–1003, 1997.CrossRefGoogle Scholar
  10. 10.
    K. Popat, R.W. Picard, “Cluster-Based Probability Model and Its Application to Image and Texture Processing”, IEEE Trans. Image Process., vol. 6, no. 2, 268–284, 1997.CrossRefGoogle Scholar
  11. 11.
    K. Rose, E. Gurewitz, G.C. Fox, “Constrained Clustering as an Optimization Method”, IEEE Trans. Pattern Anal. Machine Intell., vol. 15, 785–794, 1993.CrossRefGoogle Scholar
  12. 12.
    D.W. Scott, Multivariate Density Estimation, New York: Wiley, 1992.Google Scholar
  13. 13.
    R. Sedgewick, Algorithms in C++, New York: Addison-Wesley, 1992.Google Scholar
  14. 14.
    B.W. Silverman, Density Estimation for Statistics and Data Analysis, New York: Chapman and Hall, 1986.Google Scholar
  15. 15.
    G.R. Terrell and D.W. Scott, “Variable Density Estimation”, The Annals of Statistics, vol. 20, 1236–1265, 1992.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Dorin Comaniciu
    • 1
  • Peter Meer
    • 1
  1. 1.Department of Electrical and Computer EngineeringRutgers UniversityPiscatawayUSA

Personalised recommendations