Boundary detection in piecewise homogeneous textured images

  • Stefano Casadei
  • Sanjoy Mitter
  • Pietro Perona
Conference paper

DOI: 10.1007/3-540-55426-2_20

Volume 588 of the book series Lecture Notes in Computer Science (LNCS)
Cite this paper as:
Casadei S., Mitter S., Perona P. (1992) Boundary detection in piecewise homogeneous textured images. In: Sandini G. (eds) Computer Vision — ECCV'92. ECCV 1992. Lecture Notes in Computer Science, vol 588. Springer, Berlin, Heidelberg

Abstract

We address the problem of scale selection in texture analysis. Two different scale parameters, feature scale and statistical scale, are defined. Statistical scale is the size of the regions used to compute averages. We define the class of homogeneous random functions as a model of texture. A dishomogeneity function is defined and we prove that it has useful asymptotic properties in the limit of infinite statistical scale. We describe an algorithm for image partitioning which has performed well on piecewise homogeneous synthetic images. This algorithm is embedded in a redundant pyramid and does not require any ad-hoc information. It selects the optimal statistical scale at each location in the image.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag 1992

Authors and Affiliations

  • Stefano Casadei
    • 1
    • 2
  • Sanjoy Mitter
    • 1
    • 2
  • Pietro Perona
    • 3
    • 4
  1. 1.Massachusetts Institute of Technology 35-308CambridgeUSA
  2. 2.Scuola Normale SuperiorePisaItaly
  3. 3.California Institute of Technology 116-81PasadenaUSA
  4. 4.Università di PadovaItaly