Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Probabilistic model-based background subtraction

Krüger, Volker LU orcid ; Anderson, Jakob and Prehn, Thomas (2005) 14th Scandinavian Conference on Image Analysis, SCIA 2005 In Lecture Notes in Computer Science 3540. p.567-576
Abstract

Usually, background subtraction is approached as a pixel-based process, and the output is (a possibly thresholded) image where each pixel reflects, independent from its neighboring pixels, the likelihood of itself belonging to a foreground object. What is neglected for better output is the correlation between pixels. In this paper we introduce a model-based background subtraction approach which facilitates prior knowledge of pixel correlations for clearer and better results. Model knowledge is being learned from good training video data, the data is stored for fast access in a hierarchical manner. Bayesian propagation over time is used for proper model selection and tracking during model-based background subtraction. Bayes propagation... (More)

Usually, background subtraction is approached as a pixel-based process, and the output is (a possibly thresholded) image where each pixel reflects, independent from its neighboring pixels, the likelihood of itself belonging to a foreground object. What is neglected for better output is the correlation between pixels. In this paper we introduce a model-based background subtraction approach which facilitates prior knowledge of pixel correlations for clearer and better results. Model knowledge is being learned from good training video data, the data is stored for fast access in a hierarchical manner. Bayesian propagation over time is used for proper model selection and tracking during model-based background subtraction. Bayes propagation is attractive in our application as it allows to deal with uncertainties during tracking. We have tested our approach on suitable outdoor video data.

(Less)
Please use this url to cite or link to this publication:
author
; and
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
Image Analysis : 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005. - 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005.
series title
Lecture Notes in Computer Science
volume
3540
pages
10 pages
publisher
Springer
conference name
14th Scandinavian Conference on Image Analysis, SCIA 2005
conference location
Joensuu, Finland
conference dates
2005-06-19 - 2005-06-22
external identifiers
  • scopus:26444579313
ISSN
1611-3349
0302-9743
ISBN
978-3-540-26320-3
978-3-540-31566-7
DOI
10.1007/11499145_58
language
English
LU publication?
no
id
10db529e-88f9-4566-9b47-0771ea2717d6
date added to LUP
2019-07-08 21:26:56
date last changed
2024-01-01 16:33:37
@inproceedings{10db529e-88f9-4566-9b47-0771ea2717d6,
  abstract     = {{<p>Usually, background subtraction is approached as a pixel-based process, and the output is (a possibly thresholded) image where each pixel reflects, independent from its neighboring pixels, the likelihood of itself belonging to a foreground object. What is neglected for better output is the correlation between pixels. In this paper we introduce a model-based background subtraction approach which facilitates prior knowledge of pixel correlations for clearer and better results. Model knowledge is being learned from good training video data, the data is stored for fast access in a hierarchical manner. Bayesian propagation over time is used for proper model selection and tracking during model-based background subtraction. Bayes propagation is attractive in our application as it allows to deal with uncertainties during tracking. We have tested our approach on suitable outdoor video data.</p>}},
  author       = {{Krüger, Volker and Anderson, Jakob and Prehn, Thomas}},
  booktitle    = {{Image Analysis : 14th Scandinavian Conference, SCIA 2005, Joensuu, Finland, June 19-22, 2005.}},
  isbn         = {{978-3-540-26320-3}},
  issn         = {{1611-3349}},
  language     = {{eng}},
  month        = {{10}},
  pages        = {{567--576}},
  publisher    = {{Springer}},
  series       = {{Lecture Notes in Computer Science}},
  title        = {{Probabilistic model-based background subtraction}},
  url          = {{http://dx.doi.org/10.1007/11499145_58}},
  doi          = {{10.1007/11499145_58}},
  volume       = {{3540}},
  year         = {{2005}},
}