Practical mixtures of Gaussians with brightness monitoring

Author

Atev, Stefan ; Masoud, Osama ; Papanikolopoulos, Nikos

Author_Institution

Dept. of Comput. Sci. & Eng., Minnesota Univ., Minneapolis, MN, USA

fYear

2004

fDate

3-6 Oct. 2004

Firstpage

423

Lastpage

428

Abstract

We discuss some of the practical issues concerning the use of mixtures of Gaussians for background segmentation in outdoor scenes, including the choice of parameters. Different covariance representations and their performance impact are examined. In addition, we propose a simple, yet efficient method for coping with sudden global illumination changes based on smoothing brightness and contrast changes over time. All of the discussed methods are capable of running in real time at reasonable resolution on current generation PCs.

Keywords

Gaussian processes; covariance matrices; image resolution; image segmentation; image sequences; monitoring; traffic; video signal processing; Gaussian mixtures; background image segmentation; brightness monitoring; covariance matrices; current generation PC; illumination; image resolution; image sequences; smoothing methods; traffic monitoring; video signal processing; Brightness; Computer science; Computerized monitoring; Condition monitoring; Gaussian processes; Layout; Lighting; Personal communication networks; Surveillance; Video sequences;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Transportation Systems, 2004. Proceedings. The 7th International IEEE Conference on

Print_ISBN

0-7803-8500-4

Type

conf

DOI

10.1109/ITSC.2004.1398937

Filename

1398937