Optimization of structure elements for morphological hit-or-miss transform for building extraction from VHR airborne imagery in natural hazard areas

Chandana Dinesh Parape, Chinthaka Premachandra, Masayuki Tamura

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Template matching is a very topical issue in a wide range of imagining applications. Automated detection of features such as building roof for template matching and pattern recognition is great significance in the image processing field. In this chapter, a method which is developed optimizing the shape and size of hit-or-miss morphological filtering parameters with morphological operators is presented for building roof target detection. Morphological operations of opening and closing with constructions are applied to segmented images. Hit-or-Miss Transform (HMT) has been successfully applied for template matching in binary images. The proposed approach involves several advanced morphological operators among which an adaptive HMT with varying size and shape of the structuring elements. VHR space borne images consisting of a pre and post 2011 Pacific coast of Tohoku earthquake and the tsunami site of the Ishinomaki, Miyagi area in Japan were used. Experimental results show that the identified probability of building can reach more than 81 % by this method.

Original languageEnglish
Pages (from-to)641-650
Number of pages10
JournalInternational Journal of Machine Learning and Cybernetics
Volume6
Issue number4
DOIs
Publication statusPublished - 2015 Aug 24

Keywords

  • Airborne VHR images
  • Building extraction
  • HMT
  • Mathematical morphological profile
  • Natural hazard

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Fingerprint Dive into the research topics of 'Optimization of structure elements for morphological hit-or-miss transform for building extraction from VHR airborne imagery in natural hazard areas'. Together they form a unique fingerprint.

  • Cite this