Skip to main navigation Skip to search Skip to main content

Matching SAR images using a hierarchical constraint process

  • Massey University, Wellington

Research output: Contribution to journalConference articlepeer-review

Abstract

A relaxation labelling method was applied to match SAR images. This was done using an efficient segmentation processing chain that produces a good representation of the lines present within the SAR image. Contextual neighbourhoods for the extracted entities were established for both corner and Delaunay graphs, and a novel hierarchical approach to strengthening the constraints provided by the graph edges was presented. Results obtained revealed that by adopting a flexible approach to the establishment of local contextual neighbourhoods, the advantages of relaxation labelling can successfully be conferred to highly noise corrupted SAR images.

Original languageEnglish
Pages (from-to)60-64
Number of pages5
JournalIEE Conference Publication
Issue number410
DOIs
Publication statusPublished - 1 Oct 1995
EventProceedings of the 5th International Conference on Image Processing and its Applications - Edinburgh, UK
Duration: 4 Jul 19956 Jul 1995

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Matching SAR images using a hierarchical constraint process'. Together they form a unique fingerprint.

Cite this