Practical mixtures of Gaussians with brightness monitoring

被引：25

作者：

Atev, S ^{[1
]}

Masoud, O ^{[1
]}

Papanikolopoulos, N ^{[1
]}

机构：

[1] Univ Minnesota, Dept Comp Sci & Engn, Minneapolis, MN 55455 USA

来源：

ITSC 2004: 7TH INTERNATIONAL IEEE CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS, PROCEEDINGS | 2004年

关键词：

background segmentation; illumination monitoring; mixture models;

D O I：

10.1109/ITSC.2004.1398937

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

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.

引用

页码：423 / 428

页数：6

共 7 条

[1]

[Anonymous], 2001, P 2 EUR WORKSH ADV V

[2]

[Anonymous], PROCEEDINGS

[3]

BEVILACQUA A, 2002, 3 IAPR IND C COMP VI

[4]

Gutchess D, 2001, EIGHTH IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION, VOL I, PROCEEDINGS, P733, DOI 10.1109/ICCV.2001.937598

[5]

Lee DS, 2003, 2003 INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOL 3, PROCEEDINGS, P973

[6]

POWER W, 2002, P IM VIS COMP

[7] Computer vision algorithms for intersection monitoring [J].

Veeraraghavan, H ;

Masoud, O ;

Papanikolopoulos, NP .

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2003, 4 (02) :78-89

← 1 →