Advertisement

Signal Ensemble Classification Using Low-Dimensional Embeddings and Earth Mover’s Distance

  • Linh Lieu
  • Naoki Saito
Chapter
Part of the Applied and Numerical Harmonic Analysis book series (ANHA)

Abstract

Instead of classifying individual signals, we address classification of objects characterized by signal ensembles (i.e., collections of signals). Such necessity arises frequently in real situations: e.g., classification of video clips or object classification using acoustic scattering experiments to name a few. In particular, we propose an algorithm for classifying signal ensembles by bringing together well-known techniques from various disciplines in a novel way. Our algorithm first performs the dimensionality reduction on training ensembles using either the linear embeddings (e.g., Principal Component Analysis (PCA), Multidimensional Scaling (MDS)) or the nonlinear embeddings (e.g., the Laplacian eigenmap (LE), the diffusion map (DM)). After embedding training ensembles into a lower-dimensional space, our algorithm extends a given test ensemble into the trained embedding space, and then measures the “distance” between the test ensemble and each training ensemble in that space, and classify it using the nearest neighbor method. It turns out that the choice of this ensemble distance measure is critical, and our algorithm adopts the so-called Earth Mover’s Distance (EMD), a robust distance measure successfully used in image retrieval and image registration. We will demonstrate the performance of our algorithm using two real examples: classification of underwater objects using multiple sonar waveforms; and classification of video clips of digit-speaking lips. This article also provides a concise review on the several key concepts in statistical learning such as PCA, MDS, LE, DM, and EMD as well as the practical issues including how to tune parameters, which will be useful for the readers interested in numerical experiments.

Keywords

Principal Component Analysis Dimensionality Reduction Video Frame Hausdorff Distance Spectral Cluster 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Notes

Acknowledgements

This work was partially supported by the ONR grants N00014-06-1-0615, N00014-07-1-0166, N00014-09-1-0041, N00014-09-1-0318, the NSF grant DMS-0410406, and the NSF VIGRE grants DMS-0135345, DMS-0636297. A preliminary version of a subset of the material in this article was presented at the SPIE Wavelets XII Conference, in San Diego, in August 2007 [ 22]. We thank Dr. Quyen Huynh and Dr. Joe Lopes of NSWC-PC for providing the experimental sonar data. We have used the SPECTRAL Toolbox version 0.1 distributed on the web by Dr. Guido Sanguinetti and Dr. Jonathan Laidler to compute Elongated Kmeans. Finally, we would like to thank Dr. Bradley Marchand for his help in processing the sonar data used in the numerical experiments and Mr. Julien van Hout for his help in numerical experiments on underwater object classification.

References

  1. 1.
    Basseville, M.: Distance measures for signal processing and pattern recognition. Signal Processing 18(4), 349–369 (1989)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Belkin, M., Niyogi, P.: Laplacian eigenmaps and spectral techniques for embedding and clustering. In: T.K. Leen, T.G. Dietterich, V. Tresp (eds.) Advances in Neural Information Processing Systems, vol. 13, pp. 585–591. The MIT Press, Cambridge, MA (2001)Google Scholar
  3. 3.
    Belkin, M., Niyogi, P.: Laplacian eigenmaps for dimensionality reduction and data representation. Neural Computation 15(6), 1373–1396 (2003)MATHCrossRefGoogle Scholar
  4. 4.
    Belkin, M., Niyogi, P.: Towards a theoretical foundation for Laplacian-based manifold methods. J. Comput. Syst. Sci. 74(8), 1289–1308 (2008)MathSciNetMATHCrossRefGoogle Scholar
  5. 5.
    Bengio, Y., Paiement, J.F., Vincent, P., Delalleau, O., Roux, N.L., Ouimet, M.: Out-of-sample extensions for LLE, Isomap, MDS, Eigenmaps, and spectral clustering. In: S. Thrun, L.K. Saul, B. Schölkopf (eds.) Advances in Neural Information Processing Systems, vol. 16, pp. 177–184. The MIT Press, Cambridge, MA (2004)Google Scholar
  6. 6.
    Bıyıkoğlu, T., Leydold, J., Stadler, P.F.: Laplacian Eigenvectors of Graphs, Lecture Notes in Mathematics, vol. 1915. Springer-Verlag, New York (2007)Google Scholar
  7. 7.
    Borg, I., Groenen, P.J.F.: Modern Multidimensional Scaling: Theory and Applications, 2nd edn. Springer, New York (2005)MATHGoogle Scholar
  8. 8.
    Chung, F.R.K.: Spectral Graph Theory. No. 92 in CBMS Regional Conference Series in Mathematics. Amer. Math. Soc., Providence, RI (1997)Google Scholar
  9. 9.
    Coifman, R.R., Lafon, S.: Diffusion maps. Applied and Computational Harmonic Analysis 21(1), 5–30 (2006)MathSciNetMATHCrossRefGoogle Scholar
  10. 10.
    Coifman, R.R., Lafon, S.: Geometric harmonics. Applied and Computational Harmonic Analysis 21(1), 32–52 (2006)Google Scholar
  11. 11.
    Cover, T.M., Hart, P.: Nearest neighbor pattern classification. IEEE Trans. Inform. Theory IT-13, 21–27 (1967)Google Scholar
  12. 12.
    Donoho, D.L., Grimes, C.: Hessian eigenmaps: Locally linear embedding technique techniques for high-dimensional data. P. Natl. Acad. Sci. USA 100(10), 5591–5596 (2003)MathSciNetMATHCrossRefGoogle Scholar
  13. 13.
    Evans, L.C.: Partial differential equations and Monge-Kantorovich mass transfer. Tech. rep., Dept. Math., Univ. California, Berkeley (2001)Google Scholar
  14. 14.
    Fowlkes, C., Belongie, S., Chung, F., Malik, J.: Spectral grouping using the Nyström method. IEEE Trans. Pattern Anal. Machine Intell. 26(2), 214–225 (2004)CrossRefGoogle Scholar
  15. 15.
    Gower, J.C.: Some distance properties of latent root and vector methods used in multivariate analysis. Biometrika 53(3/4), 325–338 (1966)MathSciNetMATHCrossRefGoogle Scholar
  16. 16.
    Haker, S., Zhu, L., Tannenbaum, A., Angenent, S.: Optimal mass transport for registration and warping. Intern. J. Comput. Vision 60(3), 225–240 (2004)CrossRefGoogle Scholar
  17. 17.
    Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd edn. Springer-Verlag (2009)Google Scholar
  18. 18.
    Kirby, M., Sirovich, L.: Application of the Karhunen–Loève procedure for the characterization of human faces. IEEE Trans. Pattern Anal. Machine Intell. 12(1), 103–108 (1990)CrossRefGoogle Scholar
  19. 19.
    Lafon, S., Keller, Y., Coifman, R.R.: Data fusion and multicue data matching by diffusion maps. IEEE Trans. Pattern Anal. Machine Intell. 28(11), 1784–1797 (2006)CrossRefGoogle Scholar
  20. 20.
    Lafon, S., Lee, A.B.: Diffusion maps and coarse-graining: A unified framework for dimensionality reduction, graph partitioning and data set parameterization. IEEE Trans. Pattern Anal. Machine Intell. 28(9), 1393–1403 (2006)CrossRefGoogle Scholar
  21. 21.
    Lafon, S.S.: Diffusion maps and geometric harmonics. Ph.D. thesis, Dept. Math., Yale Univ. (2004). Downloadable from http://www.math.yale.edu/~sl349
  22. 22.
    Lieu, L., Saito, N.: Automated discrimination of shapes in high dimensions. In: D. Van De Ville, V.K. Goyal, M. Papadakis (eds.) Wavelets XII, Proc. SPIE 6701 (2007). Paper # 67011VGoogle Scholar
  23. 23.
    von Luxburg, U.: A tutorial on spectral clustering. Stat. Comput. 17(4), 395–416 (2007)MathSciNetCrossRefGoogle Scholar
  24. 24.
    van der Maaten, L.J.P., Postma, E.O., van den Herik, H.J.: Dimensionality reduction: A comparative review. Technical Report TiCC-TR 2009-005, Tilburg Centre for Creative Computing, Tilburg Univ. (2009)Google Scholar
  25. 25.
    Marchand, B.: Local signal analysis for classification. Ph.D. thesis, Dept. Math., Univ. California, Davis (2010)Google Scholar
  26. 26.
    Marchand, B., Saito, N., Xiao, H.: Classification of objects in synthetic aperture sonar images. In: Proc. 14th IEEE Workshop on Statistical Signal Processing, pp. 433–437. IEEE (2007)Google Scholar
  27. 27.
    Mardia, K.V., Kent, J.T., Bibby, J.M.: Multivariate Analysis. Academic Press, San Diego, CA (1979)MATHGoogle Scholar
  28. 28.
    Murase, H., Nayar, S.K.: Visual learning and recognition of 3d objects from appearance. Intern. J. Comput. Vision 14(1), 5–24 (1995)CrossRefGoogle Scholar
  29. 29.
    Nesbitt, C.L., Lopes, J.L.: Subcritical detection of an elongated target buried under a rippled interface. In: Oceans ’04, MTS/IEEE Techno-Ocean ’04, vol. 4, pp. 1945–1952 (2004)Google Scholar
  30. 30.
    Ng, A.Y., Jordan, M.I., Weiss, Y.: On spectral clustering: Analysis and an algorithm. In: T. Dietterich, S. Becker, Z. Ghahramani (eds.) Advances in Neural Information Processing Systems, vol. 14, pp. 849–856. The MIT Press, Cambridge, MA (2002)Google Scholar
  31. 31.
    Patterson, E.K., Gurbuz, S., Tufekci, Z., Gowdy, J.N.: Moving-talker, speaker-independent feature study, and baseline results using the CUAVE multimodal speech corpus. EURASIP J. Appl. Signal Process. 11, 1189–1201 (2002)Google Scholar
  32. 32.
    Roweis, S.T., Saul, L.K.: Nonlinear dimensionality reduction by locally linear embedding. Science 290(5500), 2323–2326 (2000)CrossRefGoogle Scholar
  33. 33.
    Rubner, Y., Tomasi, C.: Perceptual Metrics for Image Database Navigation. Kluwer Academic Publishers, Boston, MA (1999)Google Scholar
  34. 34.
    Rubner, Y., Tomasi, C., Guibas, L.J.: The Earth Mover’s Distance as a metric for image retrieval. Intern. J. Comput. Vision 40(2), 99–121 (2000)MATHCrossRefGoogle Scholar
  35. 35.
    Saito, N.: Image approximation and modeling via least statistically dependent bases. Pattern Recognition 34, 1765–1784 (2001)MATHCrossRefGoogle Scholar
  36. 36.
    Saito, N., Coifman, R.R.: Local discriminant bases and their applications. J. Math. Imaging Vis. 5(4), 337–358 (1995). Invited paperGoogle Scholar
  37. 37.
    Saito, N., Coifman, R.R., Geshwind, F.B., Warner, F.: Discriminant feature extraction using empirical probability density estimation and a local basis library. Pattern Recognition 35(12), 2841–2852 (2002)MATHCrossRefGoogle Scholar
  38. 38.
    Saito, N., Woei, E.: Analysis of neuronal dendrite patterns using eigenvalues of graph Laplacians. JSIAM Letters 1, 13–16 (2009). Invited paperGoogle Scholar
  39. 39.
    Sanguinetti, G., Laidler, J., Lawrence, N.D.: Automatic determination of the number of clusters using spectral algorithms. In: Proc. 15th IEEE Workshop on Machine Learning for Signal Processing, pp. 55–60 (2005)Google Scholar
  40. 40.
    Zhang, X., Mersereau, R.M.: Lip feature extraction towards an automatic speechreading system. In: Proc. 2000 International Conference on Image Processing, vol. 3, pp. 226–229. IEEE (2000)Google Scholar
  41. 41.
    Zhang, Z., Zha, H.: Principal manifolds and nonlinear dimensionality reduction via tangent space alignment. SIAM J. Sci. Comput. 26(1), 313–338 (2005)MathSciNetCrossRefGoogle Scholar
  42. 42.
    Zhou, S.K., Chellappa, R.: From sample similarity to ensemble similarity – Probabilistic distance measures in reproducing kernel Hilbert space. IEEE Trans. Pattern Anal. Machine Intell. 28(6), 917–929 (2006)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  1. 1.Department of MathematicsUniversity of CaliforniaDavisUSA

Personalised recommendations