Earth Mover’s Distance-Based Local Discriminant Basis

Chapter

Abstract

Local discriminant basis (LDB) is a tool to extract useful features for signal and image classification problems. Original LDB methods rely on the time–frequency energy distribution of classes or empirical probability densities, with some information theoretic measure (such as Kullback–Leibler divergence) for feature selection. Depending on the problem, energy distributions may not provide the best information for classification. Further, training set sizes and accuracy in the computed empirical probability density functions (epdfs) may hinder the learning process. To improve these deficiencies and provide a more data adaptive algorithm, we propose the use of signatures and earth mover’s distance (EMD). Signatures and EMD provide a data adaptive statistic that is more descriptive than the distribution of energies and more robust than an epdf-based approach. In this chapter, we first review LDB and EMD and then outline how they can be incorporated into a fast EMD based LDB algorithm.We then demonstrate the capabilities of our new algorithm in comparison to both energy distribution and epdf-based LDB algorithms using four different classification problems using synthetic datasets.

Keywords

Entropy Manifold Transportation 

Notes

Acknowledgments

This research was partially supported by grants NSF DMS-0410406, ONR N00014-06-1-0615, N00014-07-1-0166, N00014-09-1-0041, and N00014-09-1-0318 as well as NSF VIGRE grant DMS-0135345. Further, support was provided from the Inhouse Laboratory Independent Research (ILIR) program at the Naval Surface Warfare Center, Panama City.

References

  1. 1.
    Basseville M (1989) Distance measures for signal processing and pattern recognition. Signal Process 18:349–369MathSciNetCrossRefGoogle Scholar
  2. 2.
    Breiman L, Friedman JH, Olshen RA, Stone C (1993) Classification and regression trees. Chapman & Hall, Inc., New YorkGoogle Scholar
  3. 3.
    Daubechies I (1992) Ten lectures on wavelets. SIAM, Philadelphia, PA,MATHCrossRefGoogle Scholar
  4. 4.
    Hastie T, Tibshirani R, Friedman J (2009) The elements of statistical learning: data mining, inference, and prediction, 2nd edn. Springer, BerlinMATHCrossRefGoogle Scholar
  5. 5.
    Hillier FS, Lieberman GJ (1995) Introduction to mathematical programming. McGraw-Hill, San FranciscoGoogle Scholar
  6. 6.
    Levina E, Bickel P (2001) The earth mover’s distance is the mallow distance: Some insights from statistics. In: Proceedings of 8th IEEE International Conference on computer vision, pp 251–256Google Scholar
  7. 7.
    Rubner Y, Tomasi C (2000) Perceptual metrics for image database navigation. Kluwer Academic Publishers, BostonGoogle Scholar
  8. 8.
    Rubner Y, Tomasi C, Guibas LJ (2000) The earth mover’s distance as a metric for image retrival. Int J Comput Vis 40:99–121MATHCrossRefGoogle Scholar
  9. 9.
    Saito N (1994) Local feature extraction and its applications using a library of bases. Ph.D thesis. Department of Mathematics, Yale UniversityGoogle Scholar
  10. 10.
    Saito N, Coifman RR (1995) Local discriminant bases and their applications. J Math Imag Vis 5:337–358MathSciNetMATHCrossRefGoogle Scholar
  11. 11.
    Saito N, Coifman RR, Geshwind FB, Warner F (2002) Discriminant feature extraction using empirical probability density estimation and a local basis library. Pattern Recogn 35:2841–2852MATHCrossRefGoogle Scholar
  12. 12.
    Scott DW (1985) Averaged shifted histograms: effective nonparametric density estimators in several dimensions. Ann Stat 13:1024–1040MATHCrossRefGoogle Scholar

Copyright information

© Springer New York 2013

Authors and Affiliations

  1. 1.Panama City DivisionNaval Surface Warfare CenterPanama CityUSA
  2. 2.Department of MathematicsUniversity of CaliforniaDavisUSA

Personalised recommendations