University of Sussex
Browse

File(s) not publicly available

Robust online video background reconstruction using optical flow and pixel intensity distribution

presentation
posted on 2023-06-07, 23:57 authored by Xiaodong Cai, Falah AliFalah Ali, E Stipidis
Obtaining a dynamically updated background reference image is an important and challenging task for video applications using background subtraction. This paper proposes a novel algorithm for online video background reconstruction. Firstly, multiple candidates of background values at each pixel are obtained by locating subintervals of stable intensity in a processing period. Then criteria based on pixel intensity distribution and local optical flows are employed to decide the most likely candidate to represent the background. For the methods utilizing the distributions of intensity values, the decision of determining the background value at a pixel position is based on the observation that the appearance time and sub-period frequency of the background is higher than non-background. An enhanced method using neighborhood optical flow information is adopted for more precise decision with slightly additional computation by identifying the events of covering and revealing of a pixel position. The experimental results show that the proposed algorithm outperforms existing adaptive mixture Gaussian background model and provides robust, efficient background image reconstruction in complex and busy environment.

History

Publication status

  • Published

ISSN

1550-3607

Publisher

IEEE

Pages

5.0

Presentation Type

  • paper

Event name

IEEE International Conference on Communications (ICC 2008)

Event location

Beijing, China

Event type

conference

ISBN

978-1-4244-2075-9

Department affiliated with

  • Engineering and Design Publications

Full text available

  • No

Peer reviewed?

  • Yes

Legacy Posted Date

2012-02-06

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC