Advertisement

Multiscale Linear Feature Extraction Based on Beamlet Transform

  • Ming Yang
  • Yuhua Peng
  • Xinhong Zhou
Part of the Lecture Notes in Control and Information Sciences book series (LNCIS, volume 345)

Abstract

Beamlet [1.] is an efficient tool for multiscale image analysis. A fast algorithm for discrete beamlet transform [2.] is proposed. It greatly reduces the complexity for computing the coordinates of pixels on beamlets, and concentrates the beamlet transform on summation of the pixel grayscale values. This paper also improves Donoho’s method of using complexity-penalized energy [1.] to extract multiscale linear features. It establishes the two-scale relationship of the maximal beamlet energy in the dyadic square, and presents a threshold-processed maximal beamlet energy algorithm which can avoid the problem of selecting penalty factor. Experimental results prove the efficiency of the method proposed.

Keywords

Line Segment Fast Algorithm Linear Feature Penalty Factor Horizontal Projection 
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.

References

  1. 1.
    David L.D., Huo X.M.: Beamlets and Multiscale Image Analysis, in In Multiscale and Multiresolution Methods, Lecture Notes in Computational Science and Engineering, 20 (2001)Google Scholar
  2. 2.
    Huo X.M., Chen J.H.: JBEAM: Multiscale Curve Coding via Beamlets. IEEE Transaction on Image Processing, 14 (2005)Google Scholar
  3. 3.
    Shi Q.F., and Zhang Y.N.: Adaptive Linear Feature Detection Based on Beamlet, Processings of the Third International Conference on Machine Learning and Cybernetics, Shanghai, (2004) 26–29Google Scholar
  4. 4.
    David L. D., Huo X.M.: Beamlet Pyramids: A New Form of Multiresolution Analysis, suited for Extracting Lines, Curves, and Objects from Very Noisy Image Data, SPIE2000, (2000)Google Scholar
  5. 5.
    David L.D., Huo X.M.: Near-Optimal Detection of Geometric Objects by Fast Multiscale Methods. IEEE Transactions on Information Theory, 51(7) (2005)Google Scholar
  6. 6.
    Bresenham J.E.: Algorithm for Computer Control of a Digital Plotter. IBM Systems Journal, 4(1) (1965) 25–30CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Ming Yang
    • 1
  • Yuhua Peng
    • 1
  • Xinhong Zhou
    • 1
  1. 1.School of Information Science and EngineeringShandong UniversityJinan, Shandong ProvincePeople’s Republic of China

Personalised recommendations