Robust background segmentation using background models for surveillance application

Tianci Huang, Jingbang Qiu, Takahiro Sakayori, Takeshi Ikenaga

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

5 Citations (Scopus)

Abstract

Gaussian Mixture Models (GMM) is a very typical method for background subtraction because it possesses a strong resistibility to repetitive background motion. However when it comes to complex environment, some unexpected situations occur, e.g., when illumination changes, gradually or quickly, segmentation is generated with a poor result. Moreover, this method is not capable of distinguishing shadows of moving objects. In this paper features of intensity and texture information are utilized to eliminate the shadow of moving objects. Integrated with modified Gaussian mixture models by redefining the update criterion, proposed algorithm is adapted to the flexible illumination environment. To validate that the proposed algorithm is robust to apply on surveillance system, we provide a metric with set of variables for evaluation, a comparison had been made between proposal and original GMM, results show the accuracy improvement of models using our updated algorithm. Averagely at least of 34.8% decrease of false alarm rate proves the quality of segmentation has been significantly enhanced and proposal is more competent and stable for outdoor surveillance applications.

Original languageEnglish
Title of host publicationProceedings of the 11th IAPR Conference on Machine Vision Applications, MVA 2009
Pages402-405
Number of pages4
Publication statusPublished - 2009
Event11th IAPR Conference on Machine Vision Applications, MVA 2009 - Yokohama
Duration: 2009 May 202009 May 22

Other

Other11th IAPR Conference on Machine Vision Applications, MVA 2009
CityYokohama
Period09/5/2009/5/22

Fingerprint

Lighting
Textures

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Huang, T., Qiu, J., Sakayori, T., & Ikenaga, T. (2009). Robust background segmentation using background models for surveillance application. In Proceedings of the 11th IAPR Conference on Machine Vision Applications, MVA 2009 (pp. 402-405)

Robust background segmentation using background models for surveillance application. / Huang, Tianci; Qiu, Jingbang; Sakayori, Takahiro; Ikenaga, Takeshi.

Proceedings of the 11th IAPR Conference on Machine Vision Applications, MVA 2009. 2009. p. 402-405.

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

Huang, T, Qiu, J, Sakayori, T & Ikenaga, T 2009, Robust background segmentation using background models for surveillance application. in Proceedings of the 11th IAPR Conference on Machine Vision Applications, MVA 2009. pp. 402-405, 11th IAPR Conference on Machine Vision Applications, MVA 2009, Yokohama, 09/5/20.
Huang T, Qiu J, Sakayori T, Ikenaga T. Robust background segmentation using background models for surveillance application. In Proceedings of the 11th IAPR Conference on Machine Vision Applications, MVA 2009. 2009. p. 402-405
Huang, Tianci ; Qiu, Jingbang ; Sakayori, Takahiro ; Ikenaga, Takeshi. / Robust background segmentation using background models for surveillance application. Proceedings of the 11th IAPR Conference on Machine Vision Applications, MVA 2009. 2009. pp. 402-405
@inproceedings{392fc75d76304d4da1fcfcb00088f6ed,
title = "Robust background segmentation using background models for surveillance application",
abstract = "Gaussian Mixture Models (GMM) is a very typical method for background subtraction because it possesses a strong resistibility to repetitive background motion. However when it comes to complex environment, some unexpected situations occur, e.g., when illumination changes, gradually or quickly, segmentation is generated with a poor result. Moreover, this method is not capable of distinguishing shadows of moving objects. In this paper features of intensity and texture information are utilized to eliminate the shadow of moving objects. Integrated with modified Gaussian mixture models by redefining the update criterion, proposed algorithm is adapted to the flexible illumination environment. To validate that the proposed algorithm is robust to apply on surveillance system, we provide a metric with set of variables for evaluation, a comparison had been made between proposal and original GMM, results show the accuracy improvement of models using our updated algorithm. Averagely at least of 34.8{\%} decrease of false alarm rate proves the quality of segmentation has been significantly enhanced and proposal is more competent and stable for outdoor surveillance applications.",
author = "Tianci Huang and Jingbang Qiu and Takahiro Sakayori and Takeshi Ikenaga",
year = "2009",
language = "English",
isbn = "9784901122092",
pages = "402--405",
booktitle = "Proceedings of the 11th IAPR Conference on Machine Vision Applications, MVA 2009",

}

TY - GEN

T1 - Robust background segmentation using background models for surveillance application

AU - Huang, Tianci

AU - Qiu, Jingbang

AU - Sakayori, Takahiro

AU - Ikenaga, Takeshi

PY - 2009

Y1 - 2009

N2 - Gaussian Mixture Models (GMM) is a very typical method for background subtraction because it possesses a strong resistibility to repetitive background motion. However when it comes to complex environment, some unexpected situations occur, e.g., when illumination changes, gradually or quickly, segmentation is generated with a poor result. Moreover, this method is not capable of distinguishing shadows of moving objects. In this paper features of intensity and texture information are utilized to eliminate the shadow of moving objects. Integrated with modified Gaussian mixture models by redefining the update criterion, proposed algorithm is adapted to the flexible illumination environment. To validate that the proposed algorithm is robust to apply on surveillance system, we provide a metric with set of variables for evaluation, a comparison had been made between proposal and original GMM, results show the accuracy improvement of models using our updated algorithm. Averagely at least of 34.8% decrease of false alarm rate proves the quality of segmentation has been significantly enhanced and proposal is more competent and stable for outdoor surveillance applications.

AB - Gaussian Mixture Models (GMM) is a very typical method for background subtraction because it possesses a strong resistibility to repetitive background motion. However when it comes to complex environment, some unexpected situations occur, e.g., when illumination changes, gradually or quickly, segmentation is generated with a poor result. Moreover, this method is not capable of distinguishing shadows of moving objects. In this paper features of intensity and texture information are utilized to eliminate the shadow of moving objects. Integrated with modified Gaussian mixture models by redefining the update criterion, proposed algorithm is adapted to the flexible illumination environment. To validate that the proposed algorithm is robust to apply on surveillance system, we provide a metric with set of variables for evaluation, a comparison had been made between proposal and original GMM, results show the accuracy improvement of models using our updated algorithm. Averagely at least of 34.8% decrease of false alarm rate proves the quality of segmentation has been significantly enhanced and proposal is more competent and stable for outdoor surveillance applications.

UR - http://www.scopus.com/inward/record.url?scp=80053168838&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80053168838&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:80053168838

SN - 9784901122092

SP - 402

EP - 405

BT - Proceedings of the 11th IAPR Conference on Machine Vision Applications, MVA 2009

ER -