Publication detail

Algorithms for Interest Points Detection

ZUKAL, M. QUI, X.

Original Title

Algorithms for Interest Points Detection

English Title

Algorithms for Interest Points Detection

Type

conference paper

Language

en

Original Abstract

This paper deals with interest point detection. From many existing interest point detectors Harris detector seems to be very promising as it achieves very good results in repeatability and it is invariant to the scale and viewing angle. A modification of the Harris detector is introduced in this work. The detection algorithm is based on Haar-like features which are computed from the integral image. The algorithm has proved to be very quick in computation, but the repeatability of the proposed detector is lower in comparison with the Harris detector.

English abstract

This paper deals with interest point detection. From many existing interest point detectors Harris detector seems to be very promising as it achieves very good results in repeatability and it is invariant to the scale and viewing angle. A modification of the Harris detector is introduced in this work. The detection algorithm is based on Haar-like features which are computed from the integral image. The algorithm has proved to be very quick in computation, but the repeatability of the proposed detector is lower in comparison with the Harris detector.

Keywords

interest points, integral image, Haar-like features, genetic algorithm, parameter optimization

RIV year

2011

Released

03.02.2011

Publisher

Vysoké učení technické v Brně

Location

Antonínská 548/1, 601 90 Brno

ISBN

978-80-214-4231-3

Book

6th International Conference on Teleinformatics 2011 - ICT 2011

Edition number

1

Pages from

163

Pages to

167

Pages count

5

BibTex 


@inproceedings{BUT36379,
  author="Martin {Zukal} and Xintao {Qui}",
  title="Algorithms for Interest Points Detection",
  annote="This paper deals with interest point detection. From many existing interest point detectors Harris detector seems to be very promising as it achieves very good results in repeatability and it is invariant to the scale and viewing angle. A modification of the Harris detector is introduced in this work. The detection algorithm is based on Haar-like features which are computed from the integral image. The algorithm has proved to be very quick in computation, but the repeatability of the proposed detector is lower in comparison with the Harris detector.",
  address="Vysoké učení technické v Brně",
  booktitle="6th International Conference on Teleinformatics 2011 - ICT 2011",
  chapter="36379",
  howpublished="electronic, physical medium",
  institution="Vysoké učení technické v Brně",
  year="2011",
  month="february",
  pages="163--167",
  publisher="Vysoké učení technické v Brně",
  type="conference paper"
}