Exponentially weighted update of histogram for background modeling reducing memory usage

Tsubasa Minematsu, Masaki Igarashi, Atsushi Shimada, Hajime Nagahara, Rin-Ichiro Taniguchi

研究成果: ジャーナルへの寄稿記事

抄録

In this paper, we propose a background model by using an exponentially weighted updating method. We realize to reduce memory usage for construction of background model. Our background model is represented as a histogram according to pixel values. Our model uses an exponential increasing weight for updating our model. In our model, recently observed pixels have a bigger influence on the background model than older ones. Therefore, our model gradually ignores the effect of old-observed value on a background model without retaining past pixel values. We apply our method to background subtraction for comparing with conventional methods using kernel density estimation. In experiments, we conformed that the detection accuracy of our background model is comparable to that of conventional methods.

元の言語英語
ページ(範囲)191-200
ページ数10
ジャーナルJournal of the Institute of Image Electronics Engineers of Japan
45
発行部数2
出版物ステータス出版済み - 1 1 2016

Fingerprint

Data storage equipment
Pixels
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Science (miscellaneous)
  • Electrical and Electronic Engineering

これを引用

@article{ed578c8289fe41e3a39ba7d9fd4843da,
title = "Exponentially weighted update of histogram for background modeling reducing memory usage",
abstract = "In this paper, we propose a background model by using an exponentially weighted updating method. We realize to reduce memory usage for construction of background model. Our background model is represented as a histogram according to pixel values. Our model uses an exponential increasing weight for updating our model. In our model, recently observed pixels have a bigger influence on the background model than older ones. Therefore, our model gradually ignores the effect of old-observed value on a background model without retaining past pixel values. We apply our method to background subtraction for comparing with conventional methods using kernel density estimation. In experiments, we conformed that the detection accuracy of our background model is comparable to that of conventional methods.",
author = "Tsubasa Minematsu and Masaki Igarashi and Atsushi Shimada and Hajime Nagahara and Rin-Ichiro Taniguchi",
year = "2016",
month = "1",
day = "1",
language = "English",
volume = "45",
pages = "191--200",
journal = "Journal of the Institute of Image Electronics Engineers of Japan",
issn = "0285-9831",
publisher = "Institute of Image Electronics Engineers of Japan",
number = "2",

}

TY - JOUR

T1 - Exponentially weighted update of histogram for background modeling reducing memory usage

AU - Minematsu, Tsubasa

AU - Igarashi, Masaki

AU - Shimada, Atsushi

AU - Nagahara, Hajime

AU - Taniguchi, Rin-Ichiro

PY - 2016/1/1

Y1 - 2016/1/1

N2 - In this paper, we propose a background model by using an exponentially weighted updating method. We realize to reduce memory usage for construction of background model. Our background model is represented as a histogram according to pixel values. Our model uses an exponential increasing weight for updating our model. In our model, recently observed pixels have a bigger influence on the background model than older ones. Therefore, our model gradually ignores the effect of old-observed value on a background model without retaining past pixel values. We apply our method to background subtraction for comparing with conventional methods using kernel density estimation. In experiments, we conformed that the detection accuracy of our background model is comparable to that of conventional methods.

AB - In this paper, we propose a background model by using an exponentially weighted updating method. We realize to reduce memory usage for construction of background model. Our background model is represented as a histogram according to pixel values. Our model uses an exponential increasing weight for updating our model. In our model, recently observed pixels have a bigger influence on the background model than older ones. Therefore, our model gradually ignores the effect of old-observed value on a background model without retaining past pixel values. We apply our method to background subtraction for comparing with conventional methods using kernel density estimation. In experiments, we conformed that the detection accuracy of our background model is comparable to that of conventional methods.

UR - http://www.scopus.com/inward/record.url?scp=85031702988&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031702988&partnerID=8YFLogxK

M3 - Article

VL - 45

SP - 191

EP - 200

JO - Journal of the Institute of Image Electronics Engineers of Japan

JF - Journal of the Institute of Image Electronics Engineers of Japan

SN - 0285-9831

IS - 2

ER -