Advanced search
1 file | 970.55 KB Add to list

Efficient foreground detection for real-time surveillance applications

(2013) ELECTRONICS LETTERS. 49(18). p.1143-1144
Author
Organization
Abstract
The problem of foreground detection in real-time video surveillance applications is addressed. Proposes is a framework, which is computationally cheap and has low memory requirements. It combines two simple processing blocks, both of which are essentially background subtraction algorithms. The main novelty of the approach is a combination of an autoregressive moving average filter with two background models having different adaptation speeds. The first model, having a lower adaptation speed, models long-term background and detects foreground objects by finding areas in the current frame which significantly differ from the proposed background model. The second model, with a higher adaptation speed, models the short-term background and is responsible for finding regions in the scene with a high foreground object activity. The final foreground detection is built by combining the outputs from these building blocks. The foreground obtained by the long-term modelling block is verified by the output of the short-term modelling block, i.e. only the objects exhibiting significant motion are detected as real foreground objects. The proposed method results in a very good foreground detection performance at a low computational cost.
Keywords
computer vision, background subtraction, FG/BG segmentation

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 970.55 KB

Citation

Please use this url to cite or link to this publication:

MLA
Grünwedel, Sebastian, et al. “Efficient Foreground Detection for Real-Time Surveillance Applications.” ELECTRONICS LETTERS, vol. 49, no. 18, 2013, pp. 1143–44, doi:10.1049/el.2013.1944.
APA
Grünwedel, S., Petrovic, N., Jovanov, L., Niño Castañeda, J., Pizurica, A., & Philips, W. (2013). Efficient foreground detection for real-time surveillance applications. ELECTRONICS LETTERS, 49(18), 1143–1144. https://doi.org/10.1049/el.2013.1944
Chicago author-date
Grünwedel, Sebastian, Nemanja Petrovic, Ljubomir Jovanov, Jorge Niño Castañeda, Aleksandra Pizurica, and Wilfried Philips. 2013. “Efficient Foreground Detection for Real-Time Surveillance Applications.” ELECTRONICS LETTERS 49 (18): 1143–44. https://doi.org/10.1049/el.2013.1944.
Chicago author-date (all authors)
Grünwedel, Sebastian, Nemanja Petrovic, Ljubomir Jovanov, Jorge Niño Castañeda, Aleksandra Pizurica, and Wilfried Philips. 2013. “Efficient Foreground Detection for Real-Time Surveillance Applications.” ELECTRONICS LETTERS 49 (18): 1143–1144. doi:10.1049/el.2013.1944.
Vancouver
1.
Grünwedel S, Petrovic N, Jovanov L, Niño Castañeda J, Pizurica A, Philips W. Efficient foreground detection for real-time surveillance applications. ELECTRONICS LETTERS. 2013;49(18):1143–4.
IEEE
[1]
S. Grünwedel, N. Petrovic, L. Jovanov, J. Niño Castañeda, A. Pizurica, and W. Philips, “Efficient foreground detection for real-time surveillance applications,” ELECTRONICS LETTERS, vol. 49, no. 18, pp. 1143–1144, 2013.
@article{4106896,
  abstract     = {{The problem of foreground detection in real-time video surveillance applications is addressed. Proposes is a framework, which is computationally cheap and has low memory requirements. It combines two simple processing blocks, both of which are essentially background subtraction algorithms. The main novelty of the approach is a combination of an autoregressive moving average filter with two background models having different adaptation speeds. The first model, having a lower adaptation speed, models long-term background and detects foreground objects by finding areas in the current frame which significantly differ from the proposed background model. The second model, with a higher adaptation speed, models the short-term background and is responsible for finding regions in the scene with a high foreground object activity. The final foreground detection is built by combining the outputs from these building blocks. The foreground obtained by the long-term modelling block is verified by the output of the short-term modelling block, i.e. only the objects exhibiting significant motion are detected as real foreground objects. The proposed method results in a very good foreground detection performance at a low computational cost.}},
  author       = {{Grünwedel, Sebastian and Petrovic, Nemanja and Jovanov, Ljubomir and Niño Castañeda, Jorge and Pizurica, Aleksandra and Philips, Wilfried}},
  issn         = {{0013-5194}},
  journal      = {{ELECTRONICS LETTERS}},
  keywords     = {{computer vision,background subtraction,FG/BG segmentation}},
  language     = {{eng}},
  number       = {{18}},
  pages        = {{1143--1144}},
  title        = {{Efficient foreground detection for real-time surveillance applications}},
  url          = {{http://doi.org/10.1049/el.2013.1944}},
  volume       = {{49}},
  year         = {{2013}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: