Motion detectio n based on background modeling and performance analysis for outdoor surveillance

Tianci Huang, Jingbang Qiu, Takahiro Sakayori, Satoshi Goto, Takeshi Ikenaga

Research output: Chapter in Book/Report/Conference proceedingConference contribution

13 Citations (Scopus)

Abstract

Real-time segmentation of moving objects in video sequences is a fundamental step for surveillance systems. One of successful methods for complex background is to use a multi-color background model per pixel. However, Common problem for this approach is that it suffers from illumination changing environment, in addition, it is incapable of removing shadows of moving objects. This paper proposed an effective scheme to improve the adaptive background model for each pixel by introducing a background training parameter into every Gaussian model, and region-based scheme is applied to judgment by utilizing both spatial and temporal information. Experimental results will be presented to validate proposed algorithm keep robustness in the situation of illumination changes, shadow can be removed in foreground mask, results shows False Alarm Rate can be reduced from 34.9% to 35.8% while the overlap varies within normal range from 0.4 to 0.6 compared with conventional Gaussian mixture model.

Original languageEnglish
Title of host publicationProceedings - 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009
Pages38-42
Number of pages5
DOIs
Publication statusPublished - 2009
Event2009 International Conference on Computer Modeling and Simulation, ICCMS 2009 - Macau
Duration: 2009 Feb 202009 Feb 22

Other

Other2009 International Conference on Computer Modeling and Simulation, ICCMS 2009
CityMacau
Period09/2/2009/2/22

Fingerprint

Lighting
Pixels
Masks
Color

Keywords

  • Background
  • False Alarm Rate
  • Gaussian mixture model (GMM)
  • Training

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Science Applications
  • Software

Cite this

Huang, T., Qiu, J., Sakayori, T., Goto, S., & Ikenaga, T. (2009). Motion detectio n based on background modeling and performance analysis for outdoor surveillance. In Proceedings - 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009 (pp. 38-42). [4797351] https://doi.org/10.1109/ICCMS.2009.15

Motion detectio n based on background modeling and performance analysis for outdoor surveillance. / Huang, Tianci; Qiu, Jingbang; Sakayori, Takahiro; Goto, Satoshi; Ikenaga, Takeshi.

Proceedings - 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009. 2009. p. 38-42 4797351.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Huang, T, Qiu, J, Sakayori, T, Goto, S & Ikenaga, T 2009, Motion detectio n based on background modeling and performance analysis for outdoor surveillance. in Proceedings - 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009., 4797351, pp. 38-42, 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009, Macau, 09/2/20. https://doi.org/10.1109/ICCMS.2009.15
Huang T, Qiu J, Sakayori T, Goto S, Ikenaga T. Motion detectio n based on background modeling and performance analysis for outdoor surveillance. In Proceedings - 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009. 2009. p. 38-42. 4797351 https://doi.org/10.1109/ICCMS.2009.15
Huang, Tianci ; Qiu, Jingbang ; Sakayori, Takahiro ; Goto, Satoshi ; Ikenaga, Takeshi. / Motion detectio n based on background modeling and performance analysis for outdoor surveillance. Proceedings - 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009. 2009. pp. 38-42
@inproceedings{2d1b9cf25a8246f1b3f8574b75ded0b5,
title = "Motion detectio n based on background modeling and performance analysis for outdoor surveillance",
abstract = "Real-time segmentation of moving objects in video sequences is a fundamental step for surveillance systems. One of successful methods for complex background is to use a multi-color background model per pixel. However, Common problem for this approach is that it suffers from illumination changing environment, in addition, it is incapable of removing shadows of moving objects. This paper proposed an effective scheme to improve the adaptive background model for each pixel by introducing a background training parameter into every Gaussian model, and region-based scheme is applied to judgment by utilizing both spatial and temporal information. Experimental results will be presented to validate proposed algorithm keep robustness in the situation of illumination changes, shadow can be removed in foreground mask, results shows False Alarm Rate can be reduced from 34.9{\%} to 35.8{\%} while the overlap varies within normal range from 0.4 to 0.6 compared with conventional Gaussian mixture model.",
keywords = "Background, False Alarm Rate, Gaussian mixture model (GMM), Training",
author = "Tianci Huang and Jingbang Qiu and Takahiro Sakayori and Satoshi Goto and Takeshi Ikenaga",
year = "2009",
doi = "10.1109/ICCMS.2009.15",
language = "English",
isbn = "9780769535623",
pages = "38--42",
booktitle = "Proceedings - 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009",

}

TY - GEN

T1 - Motion detectio n based on background modeling and performance analysis for outdoor surveillance

AU - Huang, Tianci

AU - Qiu, Jingbang

AU - Sakayori, Takahiro

AU - Goto, Satoshi

AU - Ikenaga, Takeshi

PY - 2009

Y1 - 2009

N2 - Real-time segmentation of moving objects in video sequences is a fundamental step for surveillance systems. One of successful methods for complex background is to use a multi-color background model per pixel. However, Common problem for this approach is that it suffers from illumination changing environment, in addition, it is incapable of removing shadows of moving objects. This paper proposed an effective scheme to improve the adaptive background model for each pixel by introducing a background training parameter into every Gaussian model, and region-based scheme is applied to judgment by utilizing both spatial and temporal information. Experimental results will be presented to validate proposed algorithm keep robustness in the situation of illumination changes, shadow can be removed in foreground mask, results shows False Alarm Rate can be reduced from 34.9% to 35.8% while the overlap varies within normal range from 0.4 to 0.6 compared with conventional Gaussian mixture model.

AB - Real-time segmentation of moving objects in video sequences is a fundamental step for surveillance systems. One of successful methods for complex background is to use a multi-color background model per pixel. However, Common problem for this approach is that it suffers from illumination changing environment, in addition, it is incapable of removing shadows of moving objects. This paper proposed an effective scheme to improve the adaptive background model for each pixel by introducing a background training parameter into every Gaussian model, and region-based scheme is applied to judgment by utilizing both spatial and temporal information. Experimental results will be presented to validate proposed algorithm keep robustness in the situation of illumination changes, shadow can be removed in foreground mask, results shows False Alarm Rate can be reduced from 34.9% to 35.8% while the overlap varies within normal range from 0.4 to 0.6 compared with conventional Gaussian mixture model.

KW - Background

KW - False Alarm Rate

KW - Gaussian mixture model (GMM)

KW - Training

UR - http://www.scopus.com/inward/record.url?scp=64849092663&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=64849092663&partnerID=8YFLogxK

U2 - 10.1109/ICCMS.2009.15

DO - 10.1109/ICCMS.2009.15

M3 - Conference contribution

AN - SCOPUS:64849092663

SN - 9780769535623

SP - 38

EP - 42

BT - Proceedings - 2009 International Conference on Computer Modeling and Simulation, ICCMS 2009

ER -