Department of Electrical Engineering (ISY)

View Matching with Blob Features

Per-Erik Forssén, Anders Moe
Image and Vision Computing, Canadian Robotic Vision Special Issue
Volume 27, Number 1-2, Pages 99-107
January 2009

Abstract

This article introduces a new region based feature for object recognition and image matching. In contrast to many other region based features, this one makes use of colour in the feature extraction stage. We perform experiments on the repeatability rate of the features across scale and inclination angle changes, and show that avoiding to merge regions connected by only a few pixels improves the repeatability. We introduce two voting schemes that allow us to find correspondences automatically, and compare them with respect to the number of valid correspondences they give, and their inlier ratios. We also demonstrate how the matching procedure can be applied to colour correction.

Full Paper

Article is available at Science Direct.

---

Bibtex entry

@Article{fm07,
  author = 	 {Per-Erik Forss{\'e}n and Anders Moe},
  title = 	 {View Matching with Blob Features},
  journal = 	 {Image and Vision Computing, Canadian Robotic Vision Special Issue},
  year = 	 {2009},
  OPTkey = 	 {},
  volume = 	 {27},
  number = 	 {1-2},
  pages = 	 {99-107},
  month = 	 {January},
  note = 	 {Accepted October 20, 2006},
  annote = 	 {Available on-line since November 8, 2008},
  issn =         {0262-8856},
  note =          {http://dx.doi.org/10.1016/j.imavis.2006.10.005}
}