Affine Invariant Texture Segmentation and Shape from Texture by Variational Methods

  • Authors:
  • Coloma Ballester;Manuel González

  • Affiliations:
  • Departament de Matemàtiques i Informàtica, Universitat de les Illes Balears, 07071 Palma de Mallorca, Spain. E-mail: Email: dmicbn0@ps.uib.es;Departament de Matemàtiques i Informàtica, Universitat de les Illes Balears, 07071 Palma de Mallorca, Spain. E-mail: Email: dmimgh0@ps.uib.es

  • Venue:
  • Journal of Mathematical Imaging and Vision
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

We address the problem of texture segmentation by using a novel affine invariant model. The introduction of affine invariance as a requirement for texture analysis goes beyond what is known of the human performance and also beyond the psychophysical theories. We propose to compute texture features using affine invariant intrinsic neighborhoods and affine invariant intrinsic orientation matrices. We discuss several possibilities for the definition of the channels and givecomparative experimental results where an affine invariant Mumford-Shah typeenergy functional is used to compute the multichannel affine invariantsegmentation. We prove that the method is able to retrieve faithfully thetexture regions and to recover the shape from texture information in imageswhere several textures are present. The numerical algorithm is multiscale.