Object-wise multilayer background ordering for public area surveillance

Daeyong Park, Hyeran Byun

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

9 Citations (Scopus)

Abstract

Public area is one of the most significant places which need video surveillance. However, pixel-wise adaptive background subtraction methods are disturbed by incessantly passing or temporally staying foreground due to its adaptability. In such an environment, even the initialization of background is not free from the influence of foregrounds. If the adaptability is modified carelessly for selective learning, the stability of the background model will be damaged. Adjusting or fusing the learning rate slows down the false learning rate but cannot solve the problems. In this paper, we present a multilayer background modeling algorithm for public area surveillance. We efficiently cluster regions in object-wise using spatiotemporal cohesion together with spectral similarity by comparing inputs with background layer. And we classify the clustered regions and update the multi-layer model according to the results. Using the PETS data, we show that the proposed method not only maintain the background robustly but also initialize background with stationary object detection in crowded public area.

Original languageEnglish
Title of host publication6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009
Pages484-489
Number of pages6
DOIs
Publication statusPublished - 2009 Dec 28
Event6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009 - Genova, Italy
Duration: 2009 Sep 22009 Sep 4

Other

Other6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009
CountryItaly
CityGenova
Period09/9/209/9/4

Fingerprint

Multilayers
Pixels
Object detection

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Park, D., & Byun, H. (2009). Object-wise multilayer background ordering for public area surveillance. In 6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009 (pp. 484-489). [5279624] https://doi.org/10.1109/AVSS.2009.33
Park, Daeyong ; Byun, Hyeran. / Object-wise multilayer background ordering for public area surveillance. 6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009. 2009. pp. 484-489
@inproceedings{3715ad093fe74aa8b6146602e2226e81,
title = "Object-wise multilayer background ordering for public area surveillance",
abstract = "Public area is one of the most significant places which need video surveillance. However, pixel-wise adaptive background subtraction methods are disturbed by incessantly passing or temporally staying foreground due to its adaptability. In such an environment, even the initialization of background is not free from the influence of foregrounds. If the adaptability is modified carelessly for selective learning, the stability of the background model will be damaged. Adjusting or fusing the learning rate slows down the false learning rate but cannot solve the problems. In this paper, we present a multilayer background modeling algorithm for public area surveillance. We efficiently cluster regions in object-wise using spatiotemporal cohesion together with spectral similarity by comparing inputs with background layer. And we classify the clustered regions and update the multi-layer model according to the results. Using the PETS data, we show that the proposed method not only maintain the background robustly but also initialize background with stationary object detection in crowded public area.",
author = "Daeyong Park and Hyeran Byun",
year = "2009",
month = "12",
day = "28",
doi = "10.1109/AVSS.2009.33",
language = "English",
isbn = "9780769537184",
pages = "484--489",
booktitle = "6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009",

}

Park, D & Byun, H 2009, Object-wise multilayer background ordering for public area surveillance. in 6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009., 5279624, pp. 484-489, 6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009, Genova, Italy, 09/9/2. https://doi.org/10.1109/AVSS.2009.33

Object-wise multilayer background ordering for public area surveillance. / Park, Daeyong; Byun, Hyeran.

6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009. 2009. p. 484-489 5279624.

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

TY - GEN

T1 - Object-wise multilayer background ordering for public area surveillance

AU - Park, Daeyong

AU - Byun, Hyeran

PY - 2009/12/28

Y1 - 2009/12/28

N2 - Public area is one of the most significant places which need video surveillance. However, pixel-wise adaptive background subtraction methods are disturbed by incessantly passing or temporally staying foreground due to its adaptability. In such an environment, even the initialization of background is not free from the influence of foregrounds. If the adaptability is modified carelessly for selective learning, the stability of the background model will be damaged. Adjusting or fusing the learning rate slows down the false learning rate but cannot solve the problems. In this paper, we present a multilayer background modeling algorithm for public area surveillance. We efficiently cluster regions in object-wise using spatiotemporal cohesion together with spectral similarity by comparing inputs with background layer. And we classify the clustered regions and update the multi-layer model according to the results. Using the PETS data, we show that the proposed method not only maintain the background robustly but also initialize background with stationary object detection in crowded public area.

AB - Public area is one of the most significant places which need video surveillance. However, pixel-wise adaptive background subtraction methods are disturbed by incessantly passing or temporally staying foreground due to its adaptability. In such an environment, even the initialization of background is not free from the influence of foregrounds. If the adaptability is modified carelessly for selective learning, the stability of the background model will be damaged. Adjusting or fusing the learning rate slows down the false learning rate but cannot solve the problems. In this paper, we present a multilayer background modeling algorithm for public area surveillance. We efficiently cluster regions in object-wise using spatiotemporal cohesion together with spectral similarity by comparing inputs with background layer. And we classify the clustered regions and update the multi-layer model according to the results. Using the PETS data, we show that the proposed method not only maintain the background robustly but also initialize background with stationary object detection in crowded public area.

UR - http://www.scopus.com/inward/record.url?scp=72449131587&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=72449131587&partnerID=8YFLogxK

U2 - 10.1109/AVSS.2009.33

DO - 10.1109/AVSS.2009.33

M3 - Conference contribution

AN - SCOPUS:72449131587

SN - 9780769537184

SP - 484

EP - 489

BT - 6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009

ER -

Park D, Byun H. Object-wise multilayer background ordering for public area surveillance. In 6th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2009. 2009. p. 484-489. 5279624 https://doi.org/10.1109/AVSS.2009.33