The Role of Multiresolution in Mining Massive Image Datasets

  • Imola K. Fodor
  • Chandrika Kamath
Conference paper
Part of the Lecture Notes in Computational Science and Engineering book series (LNCSE, volume 20)


Scientists are collecting data from observations and simulations at an ever increasing pace. In order to extract useful information from these massive datasets, they are turning to data mining techniques as an attractive solution approach. Data mining is an iterative and interactive process that consists of data pre-processing and pattern recognition. Pre-processing the raw data in order to transform it into a form suitable for pattern recognition is an important and timeconsuming first step. In this paper, we discuss the crucial role multiresolution techniques can play in the pre-processing of massive datasets. Using both simulated and real images, we describe our work in de-noising image data using wavelet-based multiresolution techniques. Our initial experiences show that a judicious choice of wavelet transforms, threshold selection methods, and threshold application schemes can effectively reduce the noise in the data without a significant loss of the signal.


Mean Square Error Wavelet Coefficient Radio Galaxy Wavelet Shrinkage Positive Threshold 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Anant, K., Dowla, F., Rodrigue, G.: Detection of the electrocardiogram P-wave using wavelet analysis. Proc. of the SPIE 2242 (1994) 744–749.CrossRefGoogle Scholar
  2. 2.
    Burt, P., Adelson, E.: The Laplacian pyramid as a compact image code. IEEE Trans. Commun. 31 (1983) 532–540.CrossRefGoogle Scholar
  3. 3.
    Chan, T., Zhou, H.: Adaptive ENO-wavelet transforms for discontinuous functions. UCLA Technical Report 99-21 (1999) June.Google Scholar
  4. 4.
    Chan, T., Zhou, H.: Optimal construction of wavelet coefficients using total variation regularization in image compression. UCLA Technical Report CAM 00-27 (2000) July.Google Scholar
  5. 5.
    Daubechies, I.: Orthonormal bases of compactly supported wavelets. Commun. on Pure and Applied Mathematics 41 (1988) 909–996.MathSciNetzbMATHCrossRefGoogle Scholar
  6. 6.
    Daubechies, I.: Ten Lectures on Wavelets. SIAM (1992).Google Scholar
  7. 7.
    Donoho, D. L., Johnstone, I. M.: Ideal spatial adaptation via wavelet shrinkage. Biometrika 81 (1994) 425–455.MathSciNetzbMATHCrossRefGoogle Scholar
  8. 8.
    Donoho, D. L., Johnstone, I. M.: Adapting to unknown smoothness via wavelet shrinkage. JASA 90 (1995) 1200–1224.MathSciNetzbMATHCrossRefGoogle Scholar
  9. 9.
    Fayyad, U. M., Piatetsky-Shapiro, G., Smyth, P.: The KDD process for extracting useful knowledge from volumes of data. Communications of the ACM Special Issue on Data Mining 39 (1996) 27–34.Google Scholar
  10. 10.
    Fayyad, U. M., Piatetsky-Shapiro, G., Smyth, P., Uthurusamy, R.: Advances in Knowledge Discovery and Data Mining. MIT Press, Cambridge, Mass. (1996).Google Scholar
  11. 11.
    FBI Fingerprint Image Compression Standard Web Page: Scholar
  12. 12.
    Fodor, I. K., Cantü-Paz, E., Kamath, C., Tang, N. A.: Finding bent-double radio galaxies: a case study in data mining. Computing Science and Statistics 23 (2000).Google Scholar
  13. 13.
    Fodor, I. K., Kamath, C: On de-noising images using wavelet-based statistical techniques. Manuscript in preparation (2001).Google Scholar
  14. 14.
    JPEG 2000 Standard Web Page: Scholar
  15. 15.
    LeMoigne, J.: Parallel registration of multi-sensor remotely sensed imagery using wavelet coefficients. Proc. SPIE Wavelet Applications Conference (1994) 432–443.Google Scholar
  16. 16.
    Li, H., Manjunath, B. S., Mitra, S. K.: Multisensor image fusion using the wavelet transform. Proc. First International Conference on Image Processing (1994) 51–55.Google Scholar
  17. 17.
    Ma, W. Y., Manjunath, B. S.: A comparison of wavelet transform features for texture image annotation. Proc. Second International Conference on Image Processing (1995) 256–259.Google Scholar
  18. 18.
    Malladi, R., Sethian, J.: A unified approach to noise removal, image enhancement, and shape recovery. IEEE Trans. Image Processing 5 (1996) 1154–1168.CrossRefGoogle Scholar
  19. 19.
    Mallat, S.: Multiresolution approximation and wavelet orthonormal bases of L2. Trans. American Mathematical Society 315 (1989) 69–87.MathSciNetzbMATHGoogle Scholar
  20. 20.
    Meng, Q., Thompson, W., Flachs, G., Jordan, J.: Wavelet transform application in human face recognition. Proc. of the SPIE 3068 (1997) 124–135.CrossRefGoogle Scholar
  21. 21.
    MPEG-4 Standard Web Page: Scholar
  22. 22.
    Ogden, R. T.: Essential Wavelets for Statistical Applications and Data Analysis. Birkhäuser (1997).Google Scholar
  23. 23.
    Sapphire Web Page: Sapphire: Large-scale Data Mining and Pattern Recognition. Scholar
  24. 24.
    Starck, J.-L., Murtagh, F., Bijaoui, A.: Image and Data Analysis: The Multiscale Approach. Cambridge University Press (1998).Google Scholar
  25. 25.
    Strang, G., Nguyen, T.: Wavelets and Filter Banks. Wellesley-Cambridge Press (1996).Google Scholar
  26. 26.
    Uhl, A.: Wavelets: adaptive and parallel methods in image coding and signal processing. PhD thesis, University of Salzburg (1996).Google Scholar
  27. 27.
    Umbaugh, S.: Computer Vision and Image Processing: A Practical Approach using CVIPtools. Prentice Hall (1998).Google Scholar
  28. 28.
    Unser M., Aldroubi, A.: A review of wavelets in biomédical applications. Proc. of the IEEE 84 (1996) 626–638.CrossRefGoogle Scholar
  29. 29.
    Vidakovic, B.: Statistical Modeling by Wavelets. Wiley Series in Probability and Statistics. John Wiley & Sons (1999).Google Scholar
  30. 30.
    Weeks, A.: Fundamentals of Electronic Image Processing. SPIE Optical Engineering Press and IEEE Press (1996).Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Imola K. Fodor
    • 1
  • Chandrika Kamath
    • 1
  1. 1.Center for Applied Scientific ComputingLawrence Livermore National LaboratoryLivermoreUSA

Personalised recommendations