Graph edit distance with node splitting and merging, and its application to diatom identification

  • Authors:
  • R. Ambauen;S. Fischer;Horst Bunke

  • Affiliations:
  • Department of Computer Science, University of Bern, Bern, Switzerland;Department of Computer Science, University of Bern, Bern, Switzerland;Department of Computer Science, University of Bern, Bern, Switzerland

  • Venue:
  • GbRPR'03 Proceedings of the 4th IAPR international conference on Graph based representations in pattern recognition
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

A new graph matching scheme based on an extended set of edit operations, which include the splitting and merging of nodes, is proposed in this paper. This scheme is useful in applications where the nodes of the considered graphs represent regions extracted by some segmentation procedure from an image. To demonstrate the feasibility of the proposed method, its application to the automatic identification of diatoms (unicellular algae) is described.