Publication detail
VISUAL SLAM BASED ON PHASE CORRELATION AND PARTICLE FILTERS
RŮŽIČKA, M. MAŠEK, P.
Czech title
VISUAL SLAM BASED ON PHASE CORRELATION AND PARTICLE FILTERS
English title
VISUAL SLAM BASED ON PHASE CORRELATION AND PARTICLE FILTERS
Type
conference paper
Language
en
Original abstract
This paper deals with design of visual SLAM method, which is based on phase correlation and particle filters. This method can be used for localization of autonomous mobile robots inside of buildings. The method contains two parts. The first one is mapping of environ-ment, where the mobile robot operates. For this purpose was used phase correlation and images stitching method. The second one is localization problem, which was solved by particle filters, where a particles weights re-sampling was realized by phase correlation image processing method as well. Localization uses the map, which was created by phase correlation and stitching method.
Czech abstract
Tento článek se zabývá návrhem vizuální metodu SLAM, který je založen na fázové korelaci a částicových filtrů. Tato metoda může být použita pro lokalizaci mobilních autonomních robotů uvnitř budov. Metoda obsahuje dvě části. První z nich je mapování, kde mobilní robot provozuje životním-prostředí,. K tomuto účelu byla použita fáze korelaci a obrázky prošíváním metodu. Druhý z nich je lokalizace problém, který byl vyřešen filtry pevných částic, kde se částice hmotnost re-vzorkování byl realizován fáze srovnávací metodou zpracování obrazu stejně. Lokalizace využívá mapy, který byl vytvořen fázové korelace a metody šití.
English abstract
This paper deals with design of visual SLAM method, which is based on phase correlation and particle filters. This method can be used for localization of autonomous mobile robots inside of buildings. The method contains two parts. The first one is mapping of environ-ment, where the mobile robot operates. For this purpose was used phase correlation and images stitching method. The second one is localization problem, which was solved by particle filters, where a particles weights re-sampling was realized by phase correlation image processing method as well. Localization uses the map, which was created by phase correlation and stitching method.
Keywords in English
Phase correlation, particle filters, visual SLAM, localization, mapping, stitching, computer vision, image processing, mobile robot, discrete Fourier transform, inverse discrete Fourier transform and Hanning window.
RIV year
2015
Released
23.06.2015
Publisher
Springer
Location
Switzerland
ISBN
978-3-319-19823-1
ISSN
2194-5357
Book
MENDEL 2015, Recent Advances in Soft Computing
Volume
378
Edition number
378
Pages from–to
353–362
Pages count
10
BIBTEX
@inproceedings{BUT115145,
author="Michal {Růžička} and Petr {Mašek},
title="VISUAL SLAM BASED ON PHASE CORRELATION AND PARTICLE FILTERS",
booktitle="MENDEL 2015, Recent Advances in Soft Computing",
year="2015",
volume="378",
month="June",
pages="353--362",
publisher="Springer",
address="Switzerland",
isbn="978-3-319-19823-1",
issn="2194-5357"
}