Vanishing points in point-to-line mappings and other line parameterizations

  • Authors:
  • Jiří Havel;Adam Herout;MarkéTa Dubská

  • Affiliations:
  • Graph@FIT, Faculty of Information Technology, Brno University of Technology, Boetchova 2, 612 66 Brno, Czech Republic;Graph@FIT, Faculty of Information Technology, Brno University of Technology, Boetchova 2, 612 66 Brno, Czech Republic;Graph@FIT, Faculty of Information Technology, Brno University of Technology, Boetchova 2, 612 66 Brno, Czech Republic

  • Venue:
  • Pattern Recognition Letters
  • Year:
  • 2013

Quantified Score

Hi-index 0.10

Visualization

Abstract

Some variants of the Hough transform can be used for detecting vanishing points and groups of concurrent lines. This article addresses a common misconception that in the polar line parameterization the vanishing point is represented by a line. The numerical error caused by this inaccuracy is then estimated. The article studies in detail point-to-line-mappings (PTLMs) - a class of line parameterizations which have the property that the vanishing point is represented by a line (and thus can be easily searched for). When a PTLM parameterization is used for the straight line detection by the Hough transform, a pair or a triplet of complementary PTLMs has to be used in order to obtain a limited Hough space. The complementary pairs and triplets of PTLMs are formalized and discussed in this article.