Advertisement

Robust Shape Retrieval through a Novel Statistical Descriptor

  • Tuantuan Wang
  • Tong Lu
  • Wenyin Liu
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6297)

Abstract

We propose a novel statistical descriptor, Multiple References Histogram Matrix (MRHM), for robust shape retrieval, especially for degraded shape images. For each shape image, MRHM first generates uniform grids and filters noises in each grid by line Hough transformations and curve-fitting transformations. Then MRHM selects a reference for each grid and calculates its local distribution between the reference point and the shape pixels. Finally, all the local distributions are integrated into a global distribution matrix for matching symbols. Experimental results on the MPEG-7 Shape Silhouette Database and the GREC2005 Shape Database show that the proposed method’s recognition rate for degraded shape images is greatly improved over a recent method (SFHM).

Keywords

shape matching descriptor symbol recognition 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Huang, X., Gu, J., Wu, Y.: A Constraint Approach to Multifont Chinese Character Recognition. IEEE TPAMI 15, 838–843 (1993)Google Scholar
  2. 2.
    Ah-Soon, C., Tombre, K.: Architectural Symbol Recognition Using a Network of Constraints. Pattern Recognition Letters 22(2), 231–248 (2001)zbMATHCrossRefGoogle Scholar
  3. 3.
    Valveny, E., Matri, E.: A Model for Image Generation and Symbol Recognition through the Deformation of Linear Shapes. Pattern Recognition Letters 24, 2857–2867 (2003)CrossRefGoogle Scholar
  4. 4.
    Su, Y.: Symbol Recognition via Statistical Integration of Pixel-level Constraint Histograms: A New Descriptor. IEEE TPAMI 27(2), 278–281 (2005)Google Scholar
  5. 5.
    Belongie, S., Malik, J., Puzhicha, J.: Shape Matching and Object Recognition Using Shape Contexts. IEEE TPAMI 24(4), 509–522 (2002)Google Scholar
  6. 6.
    Ling, H., Jacobs, D.W.: Shape Classification Using Inner-Distance. IEEE TPAMI 29(2), 286–299 (2007)Google Scholar
  7. 7.
    Zhang, W., Liu, W.Y., Zhang, K.: Symbol Recognition with Kernel Density Matching. IEEE TPAMI 25(12), 2020–2024 (2006)Google Scholar
  8. 8.
    Zhang, J., Liu, W.Y.: A Pixel-level Statistical Structural Descriptor for Shape Measure and Recognition. In: ICDAR 2009, Spain, pp. 386–390 (2009)Google Scholar
  9. 9.
    Xu, C., Liu, J., Tang, X.: 2D Shape Matching by Contour Flexibility. IEEE TPAMI 31(1), 180–186 (2009)Google Scholar
  10. 10.
  11. 11.
    Common, P.: Independent Component Analysis, a New Concept. Signal Processing 36, 287–314 (1994)CrossRefGoogle Scholar
  12. 12.
    Hyvärinen, A., Oja, E.: Independent Component Analysis: Algorithms and Application. Neural Networks 13(4-5), 411–430 (2000)CrossRefGoogle Scholar
  13. 13.
    Song, J.Q., Su, F., et al.: Line net global vectorization: an algorithm and its performance evaluation. In: CVPR 2000, vol. (2), pp. 383–388 (2000)Google Scholar
  14. 14.

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Tuantuan Wang
    • 1
    • 3
  • Tong Lu
    • 1
  • Wenyin Liu
    • 1
    • 2
  1. 1.State Key Lab for Novel Software TechnologyNanjing UniversityNanjingChina
  2. 2.Department of Computer ScienceCity University of Hong KongHong Kong SARChina
  3. 3.Jiangyin Institute of Information Technology of Nanjing UniversityChina

Personalised recommendations