Robust foreground detection

A fusion of masked greyworld, probabilistic gradient information and extended conditional random field approach

Mohd Asyraf Zulkifley, Bill Moran, David Rawlinson

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Foreground detection has been used extensively in many applications such as people counting, traffic monitoring and face recognition. However, most of the existing detectors can only work under limited conditions. This happens because of the inability of the detector to distinguish foreground and background pixels, especially in complex situations. Our aim is to improve the robustness of foreground detection under sudden and gradual illumination change, colour similarity issue, moving background and shadow noise. Since it is hard to achieve robustness using a single model, we have combined several methods into an integrated system. The masked grey world algorithm is introduced to handle sudden illumination change. Colour co-occurrence modelling is then fused with the probabilistic edge-based background modelling. Colour co-occurrence modelling is good in filtering moving background and robust to gradual illumination change, while an edge-based modelling is used for solving a colour similarity problem. Finally, an extended conditional random field approach is used to filter out shadow and afterimage noise. Simulation results show that our algorithm performs better compared to the existing methods, which makes it suitable for higher-level applications.

Original languageEnglish
Pages (from-to)5623-5649
Number of pages27
JournalSensors (Switzerland)
Volume12
Issue number5
DOIs
Publication statusPublished - May 2012

Fingerprint

Fusion reactions
Color
fusion
Lighting
color
gradients
illumination
afterimages
Noise
occurrences
Afterimage
Detectors
detectors
background noise
Face recognition
traffic
counting
Pixels
pixels
filters

Keywords

  • Colour co-occurrence
  • Colour constancy
  • Conditional random field
  • Edge-based modelling
  • Foreground detection
  • Gaussian modelling
  • Shadow removal

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Atomic and Molecular Physics, and Optics
  • Analytical Chemistry
  • Biochemistry

Cite this

Robust foreground detection : A fusion of masked greyworld, probabilistic gradient information and extended conditional random field approach. / Zulkifley, Mohd Asyraf; Moran, Bill; Rawlinson, David.

In: Sensors (Switzerland), Vol. 12, No. 5, 05.2012, p. 5623-5649.

Research output: Contribution to journalArticle

@article{fe60424ad2c344d7b06fbb5f556bd51c,
title = "Robust foreground detection: A fusion of masked greyworld, probabilistic gradient information and extended conditional random field approach",
abstract = "Foreground detection has been used extensively in many applications such as people counting, traffic monitoring and face recognition. However, most of the existing detectors can only work under limited conditions. This happens because of the inability of the detector to distinguish foreground and background pixels, especially in complex situations. Our aim is to improve the robustness of foreground detection under sudden and gradual illumination change, colour similarity issue, moving background and shadow noise. Since it is hard to achieve robustness using a single model, we have combined several methods into an integrated system. The masked grey world algorithm is introduced to handle sudden illumination change. Colour co-occurrence modelling is then fused with the probabilistic edge-based background modelling. Colour co-occurrence modelling is good in filtering moving background and robust to gradual illumination change, while an edge-based modelling is used for solving a colour similarity problem. Finally, an extended conditional random field approach is used to filter out shadow and afterimage noise. Simulation results show that our algorithm performs better compared to the existing methods, which makes it suitable for higher-level applications.",
keywords = "Colour co-occurrence, Colour constancy, Conditional random field, Edge-based modelling, Foreground detection, Gaussian modelling, Shadow removal",
author = "Zulkifley, {Mohd Asyraf} and Bill Moran and David Rawlinson",
year = "2012",
month = "5",
doi = "10.3390/s120505623",
language = "English",
volume = "12",
pages = "5623--5649",
journal = "Sensors (Switzerland)",
issn = "1424-8220",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "5",

}

TY - JOUR

T1 - Robust foreground detection

T2 - A fusion of masked greyworld, probabilistic gradient information and extended conditional random field approach

AU - Zulkifley, Mohd Asyraf

AU - Moran, Bill

AU - Rawlinson, David

PY - 2012/5

Y1 - 2012/5

N2 - Foreground detection has been used extensively in many applications such as people counting, traffic monitoring and face recognition. However, most of the existing detectors can only work under limited conditions. This happens because of the inability of the detector to distinguish foreground and background pixels, especially in complex situations. Our aim is to improve the robustness of foreground detection under sudden and gradual illumination change, colour similarity issue, moving background and shadow noise. Since it is hard to achieve robustness using a single model, we have combined several methods into an integrated system. The masked grey world algorithm is introduced to handle sudden illumination change. Colour co-occurrence modelling is then fused with the probabilistic edge-based background modelling. Colour co-occurrence modelling is good in filtering moving background and robust to gradual illumination change, while an edge-based modelling is used for solving a colour similarity problem. Finally, an extended conditional random field approach is used to filter out shadow and afterimage noise. Simulation results show that our algorithm performs better compared to the existing methods, which makes it suitable for higher-level applications.

AB - Foreground detection has been used extensively in many applications such as people counting, traffic monitoring and face recognition. However, most of the existing detectors can only work under limited conditions. This happens because of the inability of the detector to distinguish foreground and background pixels, especially in complex situations. Our aim is to improve the robustness of foreground detection under sudden and gradual illumination change, colour similarity issue, moving background and shadow noise. Since it is hard to achieve robustness using a single model, we have combined several methods into an integrated system. The masked grey world algorithm is introduced to handle sudden illumination change. Colour co-occurrence modelling is then fused with the probabilistic edge-based background modelling. Colour co-occurrence modelling is good in filtering moving background and robust to gradual illumination change, while an edge-based modelling is used for solving a colour similarity problem. Finally, an extended conditional random field approach is used to filter out shadow and afterimage noise. Simulation results show that our algorithm performs better compared to the existing methods, which makes it suitable for higher-level applications.

KW - Colour co-occurrence

KW - Colour constancy

KW - Conditional random field

KW - Edge-based modelling

KW - Foreground detection

KW - Gaussian modelling

KW - Shadow removal

UR - http://www.scopus.com/inward/record.url?scp=84861549011&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861549011&partnerID=8YFLogxK

U2 - 10.3390/s120505623

DO - 10.3390/s120505623

M3 - Article

VL - 12

SP - 5623

EP - 5649

JO - Sensors (Switzerland)

JF - Sensors (Switzerland)

SN - 1424-8220

IS - 5

ER -